MENTI Synced 2026-07-15 · Build #437

MENTI Research Status — 2026-05-28

Articles

Accepted

Under Review

Data Done

Open Tasks

Hard Deadlines <90d

Live Surveys

162

Activity Log

This Week

1. #50 · OVERDUE

Article 14 Clalit - Clalit sent an email (2026-06-15): update...

Elad + Yuval · 2026-06-30

2. #52 · TODAY

Recruit Research Assistant (RA) for MENTI - flyer published 2...

Elad · 2026-07-15

3. #36 · NO DEADLINE

MENTI website: add contact form / details submission

TBD (web)

4. #49 · NO DEADLINE

Article 3 Evelyn - build 6 multi-session prompts + deploy to ...

Yuval

5. #53 · NO DEADLINE

Article 9 (CPE vs flat-prompt LIWC) - deploy CPE Alice on Ces...

Elad + Yuval

Nature Awards 2026: SUBMITTED 2026-05-27 (Health & Life Sciences)

#50 Article 14 Clalit - Clalit sent an email (2026-06-15): update the Clalit proposal accordingly (owner Elad) Elad + Yuval 2026-06-30 ▸

#52 Recruit Research Assistant (RA) for MENTI - flyer published 2026-06-26; track applications/responses Elad 2026-07-15 ▸

#12 Yoav Article 5 - Active recruitment + weekly monitoring (OSF fz8xm live 2026-05-28) Elad — ▸

#24 Nature Awards 2026 outcome - track decision (SUBMITTED via Nature portal 2026-05-27; notification expected per Nature programme timeline; owner Elad) Elad — ▸

#25 Dashboard health audit (weekly) Claude/menti-agent (Telegram bot) — ▸

#36 MENTI website: add contact form / details submission TBD (web) — ▸

#47 Article 17 Hili pilot - Elad reviews preliminary report + decide on powered/controlled follow-up (Year-2 plan) Elad — ▸

#49 Article 3 Evelyn - build 6 multi-session prompts + deploy to Cesura + engagement Yuval — ▸

#51 MENTI Study Studio - Yuval review + schedule walkthrough with Elad Yuval — ▸

#53 Article 9 (CPE vs flat-prompt LIWC) - deploy CPE Alice on Cesura + recruit ~16 therapists (protocol + prompts + IRB all ready) Elad + Yuval — ▸

#55 Article 13 (Prompt Compression) - launch rater recruitment + open blinded expert evaluation (platform live, stimuli + IRB ready) Elad + Yuval — ▸

#56 Article 13: add research arm - therapists converse with BOTH Sara bots (33K vs 3K) themselves + judge, beyond rating the recorded portal sessions Yuval — ▸

#57 Evelyn study: build detailed validated-questionnaire table (items, languages, codes) from the Questionnaires folder Yuval — ▸

#58 Evelyn study: consult Karen on shortening/merging questionnaires to cut participant burden Elad + Yuval — ▸

#59 Fix Dropbox sync - coordinator files arrive EMPTY for Elad (technical) Yuval — ▸

#60 Confirm Moria (overseas RA) arrival dates + plan intensive MENTI integration Yuval — ▸

#63 Article 17 (Hili): review v3 manuscript + decide Scale B status / length / author lists Elad — ▸

#68 Organize Evelyn (Article 3) questionnaires in BOTH Hebrew and English Yuval Holzman — ▸

#69 Learn Prolific + the Evelyn-on-Prolific run protocol Yuval Holzman — ▸

#70 Yuval learns (with her own Claude) to connect Claude Code to GitHub Yuval — ▸

#71 Yuval learns (with her own Claude) to build a questionnaire-building skill, based on Elad's Yuval — ▸

#74 UP big paper (12B) -- submit full study to JMIR Medical Education (npj deadline dropped) Elad — ▸

#75 UP lecture study (Article 18) -- build: questionnaire + Cesura workspace + flyer + Zohar consult Shaked (questionnaire+workspace+recruitment), Omer (deliver lecture), Elad (flyer+Zohar) — ▸

#76 Zvia Zeligman bot study — planned separate bot study (Elad + Zvia); assign article # Elad — ▸

#77 Article 15 Rina — schedule a meeting with Inbal Cohen about Rina's bot Elad — ▸

#1 MentiParent (Danny + Evelyn) Parent VERIFY ACCEPTED Verify Scientific Reports production/publication status + collect p... 2026-07-02 ▸

#2 Foster Care (Julie/Guli) Foster VERIFY UNDER REVIEW Ping Tal - reviewer silence since May; request journal status (refr... 2026-07-02 ▸

#3 Evelyn new run Parent VERIFY BUILDING Design locked (mtg 2026-06-08): 3 bots/sessions; ADD SDQ + Parentin... 2026-07-15 ▸

#4 Reimagining Clinical Training (Carol+Dweck triad) Therapists IN PREP Carol + Dafna LIVE on Elad's platform (Carol 260626; Dafna V8). Mod... 2026-07-15 ▸

#5 Socratic Yoav Therapists ACTION RECRUITING Study 2 confirmatory sample = new=1 responses 2026-06-26 onward (ex... 2026-07-02 ▸

#6 AI Rater Methodology IN PREP Expert review of prompts -> AI rating run -> calibration (unchanged... 2026-07-02 ▸

#7 Or experts FosterMethodology VERIFY IN PREP Get fresh status from Tal - expert eval stalled since April; chase ... 2026-07-02 ▸

#8 CPE / CLPsych (Lia) Methodology ACCEPTED Yuval prepares CLPsych 2026 slides + script from the locked outline... 2026-07-05 ▸

#9 CPE vs Flat Prompt LIWC Methodology READY TO DEPLOY English-session constraint noted (briefed to Karen). Deployment not... 2026-07-02 ▸

#10 Narrative Review (5-Axis) Methodology ACTION UNDER REVIEW npj technical check in progress -> peer review; decision ~4-10 wks 2026-07-07 ▸

#11 Deep Persona Methodology UNDER REVIEW Not in current focus; to revisit later (per Elad 2026-06-21). 2026-06-21 ▸

#12A UP Pilot (Paper #1) Therapists ACTION DRAFTING SUPERSEDED by the 2026-07-02 Elad+Omer+Shaked call: the old 'small ... 2026-07-15 ▸

#12B UP Within-Subjects (Paper #2) Therapists ACTION HARD DEADLINE 2026-07-31 Full-study 'big paper' (Elad first author, target JMIR Medical Educ... 2026-07-15 ▸

#13 Prompt Compression Methodology RUNNING Rating platform live; within-subjects blinded eval (10 pairs x 14 i... 2026-07-02 ▸

#14 Clalit / Dalia Parent VERIFY PROTOCOL ADVANCE the study. Blocker = the contract/undertaking: fix the NAME... 2026-07-15 ▸

#15 Rina Therapists ACTION QUESTIONNAIRE LIVE Two live surveys (NEVER pool). INTERVENTION SV_734qjklvoJkl4to: 45 ... 2026-07-15 ▸

#16 Menti Zimbardo Style Methodology STUB Stub - parked; scope when Elad raises it (reviewed 2026-07-02) 2026-07-02 ▸

#17 Hili Pilot — MentiTherapist TherapistsParent ACTION DATA COLLECTION DONE v3 manuscript AUDIT-PASSED + Word sent to Elad (USER-LOCK pending, ... 2026-06-15 ▸

#18 UP Lecture Study Therapists PLANNING NEW standalone 'lecture study' decided in the 2026-07-02 Elad+Omer+... 2026-07-15 ▸

Surveys Live

631

Total Responses

UP Unique Participants

367

UP Response Rows

Surveys to Verify

BIU N

Haifa N

Hard Deadlines

Qualtrics Live Surveys 12 surveys · as-of PILOT 2026-06-08

Label	Survey ID	Responses	As-of	Notes
Yoav (Article 5)	SV_039jMzxfdPX6yRU	337	2026-06-12	Refreshed 2026-05-28 from AI_PSYCH API. +5 since 2026-05-18 (slow CPT-cohort drip).
Julie (Article 2)	SV_eJ5kD9LnE0bfR7U	71	2026-06-12	Stale -- last known 2026-05-18. AI_PSYCH/ISF/BSF API keys returned 403 (not i...
Or experts (Article 7)	SV_6kZsW4Ky4nPndno	59	2026-06-12	Refreshed 2026-05-28 from AI_PSYCH API. Unchanged since 2026-05-18.
Danny-1 (Article 1)	SV_4ZKQWeGytSdBEbk	50	2026-06-12	Stale -- last known 2026-05-18. AI_PSYCH/ISF/BSF API keys returned 403 (not i...
ICHILOV	SV_9NoJO19hm2ACw0S	21	2026-06-12	Stale -- last known 2026-05-18. AI_PSYCH/ISF/BSF API keys returned 403 (not i...
menti_test ⚠ VERIFY	REMOVED_NOT_IN_AI_PSYCH	34	2026-05-18	ICHILOV survey not found in AI_PSYCH/ISF/BSF accounts. Check Bar-Ilan ICHILOV...
Danny-2	SV_eFGhPljShwykLB4	34	2026-06-12	Stale -- last known 2026-05-18. AI_PSYCH/ISF/BSF API keys returned 403 (not i...
Rina (Assor, side project)	SV_71cNAGWbJOvyfci	17	2026-06-12	Refreshed 2026-05-28 from AI_PSYCH API. Older/different Assor survey -- NOT A...
Lia	SV_0wXR00x6HcR8o3I	1	2026-06-12	Stale -- last known 2026-05-18. AI_PSYCH/ISF/BSF API keys returned 403 (not i...
Rina (Article 15 OFFICIAL)	SV_734qjklvoJkl4to	7	2026-06-12	Refreshed 2026-05-28 from AI_PSYCH API. +7 since 2026-05-18 -- first response...
Hili Pilot (Article 17 -- PRE)	TBD (build 2026-05-31)	0	PILOT 2026-06-08	Article 17 MentiTherapist pilot. PRE Qualtrics build owed by Task #26 due 202...
Hili Pilot (Article 17 -- POST)	TBD (build 2026-05-31)	0	PILOT 2026-06-08	Article 17 MentiTherapist pilot. POST Qualtrics build owed by Task #26 due 20...

UP — BIU site as-of 2026-05-25

Q-Pre

Q6+Post

Canonical N=32 unique participants · Sessions 1–4 all completed · Led by Elad + Omer + Shaked

UP — Haifa site as-of 2026-05-25

Q-Pre

Q6+Post

Canonical N=39 unique participants · Led by Joy Bentov · In-class sessions · 39 enrolled / 12 done / 27 follow-up

UP totals reconciliation: N=71 unique participants vs N=367 clean response rows (layer-1) vs N=363 deep-audit rows (layer-3). All valid — different purposes. Use N=71 for participant-level stats; N=363/367 for response-level (retention funnels, MLM observations).

2026-07-15 status UPDATE 162 · UP pre-reg SCOPE flag resolved (Elad 2026-07-15) ▸

Fresh pre-registration to be PREPARED FROM SCRATCH for 12B (Article 122) -- the active upstream pre-reg deliverable. Article 18 already has its own working draft (07-14 AsPredicted, not finalized/registered) and is NOT the prepared-from-scratch record. Dropped 'SCOPE=CONFIRM WITH ELAD' from Art.122/#74 and Art.18/#75. No status flips.

📄 ops.py-driven

2026-07-15 milestone UPDATE 161 · Elad 2026-07-15 program update absorbed (6 items) ▸

A14 Clalit: advance; blocker = fix contract name + add Elad as signatory, via Elad->Omer. A3 Evelyn + A4 Carol/Dafna studies + Zvia Zeligman bot study: all gated on Yuval's return (reset when back). UP 12B: built from Zohar's IPT protocol; Shaked finished code review; final joint Elad+Shaked review Thu 2026-07-16. A15 Rina: schedule Inbal meeting (task #77). Zvia Zeligman planned bot study logged (task #76, article# TBD). CAM/P1 (Hadar+Mor) = AI-TOM side project, NOT MENTI — routed to cross-project log.

📄 ops.py-driven

2026-07-15 status UPDATE 160 · UP pre-registration status corrected (12B + Article 18) ▸

Elad 2026-07-15: NO finalized/approved/registered pre-reg exists for 12B OR Article 18 -- only early working drafts (12B OSF-style v4.0/v3.4; Art.18 07-14 AsPredicted). The pre-reg is being redone FROM SCRATCH; those drafts are superseded starting points, not near-ready. Tightened next_step on Articles 122+18 and descriptions on tasks #74+#75 so no future session over-reads them. SCOPE of the fresh redo (12B only / Art.18 only / both) = flagged confirm-with-Elad. OSF/AsPredicted click stays pending Elad. No status flips.

📄 ops.py-driven

2026-07-15 status UPDATE 159 · UP (12B): absorbed 3 Elad status updates -- items now in progress ▸

Elad 2026-07-15: (1) v6.1 cleaning USER-LOCK is under joint Elad+Shaked code review right now (Shaked=second reviewer, not a blocker); (2) the 12 open Study-2 decisions + the pre-registration are in preparation by Elad/team (OSF click still pending Elad); (3) Barlow co-authorship on 12B CONFIRMED to be approached at the END, after a draft exists. Refreshed Article 122 next_step + task #74 via ops.py. No data.json status flip (all stay in_preparation). validate 12/12.

📄 ops.py-driven

2026-07-13 publications UPDATE 158 · MENTI-funded output: npj Digital Medicine Perspective submitted (grant 0008682) ▸

'I Do Not Think It Means What You Think It Means: The Many Meanings of Agent in Mental Health AI' submitted 2026-07-13 to npj Digital Medicine (Collection: Impact of Agentic AI on Care Delivery; Snapp ec65140a, technical check). Piterman/Refoua co-first; Elyoseph corresponding. Funding acknowledged: E.R. and Z.E., Israel Ministry of Innovation Science and Technology, MENTI Project grant 0008682. NOT a MENTI article-study (Agents research line, parallel to MENTI per Elad 2026-07-12) - logged here as a grant output only.

📄 ops.py-driven

2026-07-12 update UPDATE 157 · Article 15 (Rina) response-status snapshot + Inbal control-recruitment push ▸

Read-only 2026-07-12 pull (metadata only, no answer content): intervention SV_734qjklvoJkl4to = 45 real (37 completed + 8 partial; 10 dev/test recorded <=2026-06-17 excluded by date). Control SV_3CQw2i9BhfgMuAm = 4 real (1 completed + 3 partial) -- control recruitment is the bottleneck. Sent Inbal a status message with both counts + the dev/test note + a control-link push (each once, signed). No survey changes.

📄 ops.py-driven

2026-07-07 status_change UPDATE 156 · Article 10 resubmitted to npj Digital Medicine ▸

After JMIR MedEd editorial decline (too formative), Article 10 was submitted 2026-07-07 to the npj Digital Medicine collection Transforming Medical Education through AI (Snapp 59b0d136). Cover letter repositioned per Zohar (organizing synthesis around the transfer question). Technical check in progress.

📄 ops.py-driven

2026-07-06 milestone UPDATE 155 · Article 12 (UP) reshaped after 2026-07-02 Elad+Omer+Shaked call ▸

Portfolio update propagated from the UP simu project. Decisions: (1) npj DM special-issue 31-Jul-2026 HARD deadline DROPPED (all agreed unattainable). (2) Two papers reshaped -- BIG (12B) = full study, story 'sim-trained skills improve more than non-sim', Elad first author, target JMIR Medical Education (IF ~13.9; Nature Mental Health option open); SMALL = a NEW standalone lecture study (Article 18) modeled on Zohar's IPT evenings (module lecture + one bot run for practicing therapists, pre/post, Facebook recruitment ~100) -- replaces the DROPPED M3 single-sim pilot (12A). (3) Predictors/anthropomorphism paper DROPPED. (4) Cleaning code rebuilt for readability; Shaked = second reviewer who signs off. (5) Barlow co-authorship deferred until after a draft exists. Roles: Shaked pushes lecture + questionnaire + new Cesura workspace; Omer delivers lecture; Elad flyer + Zohar consult. State: created Article 18; updated 12A/12B; closed tasks #18 #20; added JMIR-submission + lecture-build tasks.

📄 ops.py-driven

2026-07-06 milestone UPDATE 154 · Article #18 added: UP Lecture Study -- UP-module lecture + AI simulation for practicing therapists (pre/post, Facebook recruitment, Zohar-IPT model) ▸

New article registered via ops.py article add. Track=therapists; folder=study/articles/18. UP lecture study/

📄 ops.py-driven

2026-07-06 status_change UPDATE 153 · UP small paper (12A) M3-pilot framing dropped; role moved to new Article 18 lecture study ▸

2026-07-02 call: the M3 single-sim pilot (course-dataset carve-out) is DROPPED (Omer: not really a pilot, weakens big-paper innovation; 06-28 decision = use ALL data). The 'small paper' is now a NEW standalone lecture study (Article 18). 12A retained for audit only.

📄 ops.py-driven

2026-07-06 status_change UPDATE 152 · UP big paper (12B) reshaped: journal -> JMIR Med Ed, npj deadline dropped ▸

2026-07-02 Elad+Omer+Shaked call: 12B is now the full-study big paper (story = sim-trained skills improve more than non-sim). Target npj DM special issue -> JMIR Medical Education (IF ~13.9); Nature Mental Health option open; npj 31-Jul-2026 hard deadline DROPPED. Elad first author; Barlow deferred post-draft.

📄 ops.py-driven

2026-07-06 status_change UPDATE 151 · Article 10 SUBMITTED to JMIR Medical Education (#106300) ▸

The five-domain narrative review 'LLM-Based Simulated Patients for Interpersonal Skills Training in Healthcare and Mental Health Education' was submitted 2026-07-06 to JMIR Medical Education (submission #106300, section AI in Medical Education). Status Unassigned = under review, awaiting editor assignment. Journal switched from target Advances in Simulation to JMIR Medical Education. Corresponding author Zohar Elyoseph.

📄 ops.py-driven

2026-07-06 status_change UPDATE 150 · Article 10: JMIR submission prep + reference library + Ross Table-1 fix ▸

Corrected manuscript (17 approved copy-edits verified; refs now 1-62 all cited, 45 studies consistent); cover letter on Zohar U-Haifa letterhead (writer-polished, Professor); suggested-reviewers doc (4 clean + Benatov COI-flagged); reference_library/ with 44 verified study PDFs + master table (md+docx); caught & fixed a real data-integrity issue - Table 1 Row 45 'Ross' held the Cross study's data (mislabeled source PDF); re-extracted from the real Ross paper (nursing, N=164, therapeutic communication) after Elad+Yuval confirmed Ross is a distinct study; body prose was already correct. Fix-list documents remaining additions (keywords/ethics/data-availability) + portal steps. Next: Elad adds those sections + Claude fills JMIR portal for Elad's submit click.

📄 ops.py-driven

2026-07-05 status_change UPDATE 149 · Article 10: JMIR Med Ed submission package prepared ▸

Multi-agent submission-readiness review (33 agents, verified): JMIR ME requirements pulled; 27 real issues confirmed (key: 46->45 in Methods; uncited refs 63-67; keywords/ethics/data-availability sections missing; ~14 copy-edits) + 2 false-positives rejected. Built: cover letter on Zohar U-Haifa letterhead (45 corrected, narrative-vs-systematic justification, self-citations, affirmative AI disclosure), suggested-reviewers doc (Asman/Amir Tal/Ophir/Lauderdale clean; Benatov COI-flagged), SUBMISSION_FIXLIST. Letterhead template saved for reuse. Next: apply fixes on Elad's OK, then fill JMIR portal (Elad clicks submit).

📄 ops.py-driven

2026-07-05 status_change UPDATE 148 · Article 8: Yuval presented at CLPsych 2026 (oral) + excellence scholarship ▸

Talk delivered 2026-07-05 with good reactions; Yuval also won an excellence scholarship for the presentation. Companion page sim link disabled post-conference (live-verified); Facebook post prepared for Elad (CPE method + Talking-with-Lia case study + built-to-help-in-wartime origin), paper + demo page links go in first comment.

📄 ops.py-driven

2026-07-02 status_change UPDATE 147 · Article 8: CLPsych companion page finalized + Lia sim link live ▸

lia-clpsych-companion updated per Elad: (1) custom wrapper fullscreen so video zoom (now up to 3x) works inside fullscreen, zoom rebuilt width-based to reach all corners; (2) availability changed to open-all-conference-day; (3) English Lia sim (app.cesura.ai/sim/8c5a5780...) wired into Open Simulation, enabled, live-verified + participant smoke-test to the consent gate; test request sent once to menti-rachav (msg 3EB0F4D7..., 23:47 IL). Full dossier: project_menti_article8_clpsych_talk.md

📄 ops.py-driven

2026-07-02 update UPDATE 146 · UP (Article 12) task reconciliation vs v6 truth ▸

Reconciled dashboard UP/Article-12 tasks against the 2026-07-02 UP truth (v6 cleaning BUILT + AUDIT-PASSED 7/7, N=76 = BIU 34 + Haifa 42 / 702 LONG, NOT user-locked; prereg v3.4 never registered + v4.0 drafted not registered). CLOSED #15 (v3.4 codebook sync -- superseded by v6 rebuild's own generated codebook) + #19 (2026-06-08 meeting -- date passed, superseded by 06-29 Elad+Shaked meeting). REFRESHED #20 (npj DM Paper 2 -- set due 2026-07-31 + v6/prereg-v4 description) + #18 (Tal qualitative -- confirmed still valid, Paper 1 arm). Refreshed 12A/12B next_step to v6 AUDIT-PASSED. FLAGGED for Elad: the 3-way meeting with Omer is not yet a dashboard task.

📄 ops.py-driven

2026-07-02 update UPDATE 145 · Super-agent promo video produced + sent to Yuval ▸

1:47 Hebrew promo about the MENTI super-agent (Fable upgrade: Timeless meeting-reading, self-healing WhatsApp bridge, live timeline, hub role). First production using GPT Image 2 frames (all 5 frames, Hebrew titles rendered correctly - video-producer agent updated with the new generator). Voice Laomedeia (Gemini TTS). QC'd frame-by-frame. Sent ONCE to Yuval Holzman via WhatsApp as document (msg 3EB0D8212E7F546C82EA74, 16:10 IL, signed Claude). File: menti/_assets/menti_superagent_video_2026-07-02/

📄 ops.py-driven

2026-07-02 update UPDATE 144 · Infrastructure overhaul + Timeless meetings integration ▸

CLAUDE.md diet 98KB->9KB (history archived verbatim); PROJECT_TIMELINE.md created at project root (single-glance iron-rule file); 27 past-TTL session logs archived; cross_project_sync_log rotated 425KB->29KB + hook 6 now collapses same-day auto-pushes; hook 5 rewired to update PROJECT_TIMELINE instead of growing CLAUDE.md; dashboard repo dead files removed; 50 root session artifacts -> _assets; drift triage: 7 evidently-done/superseded tasks closed + 4 tasks & 6 articles refreshed; menti-agent gained Timeless capability (reference_timeless_meetings.md; 238 meetings reachable, Hebrew search verified).

📄 ops.py-driven

2026-07-02 milestone UPDATE 143 · 9<->13 possible merge under consideration - leaning to keep separate for now ▸

Elad's design (WhatsApp to Karen 2026-07-01): one 3-arm within-subjects study (flat / full CPE ~33K / compressed CPE ~3K), 10 therapists x all 3, measured by LIWC + clinician richness ratings; 1v2=#9, 2v3=#13, 1v3=bonus. Elad UNDECIDED, leaning keep #9 and #13 separate (both nearly ready) + cross-pollinate cheaply. OPEN option, not a decision.

📄 ops.py-driven

2026-07-02 milestone UPDATE 142 · Karen briefed (WhatsApp) on Articles 9 + 13 + Evelyn status ▸

Deep update to Karen Yirmiya, delivered once, signed (kloud): Article 9 English-session constraint; the possible 9<->13 merge; Evelyn (Article 3) Prolific->RCT + length/usage deliberations.

📄 ops.py-driven

2026-07-02 milestone UPDATE 141 · Yoav (Art.5) first Cesura run: 20 started / 8 completed / 12 partial ▸

Fri 2026-06-26 (Danny Derby group): 20 started -> 8 completed (100%) / 12 not finished (6 at ~4%, 6 engaged at 26-35% missing only the POST). All 8 completers new=1. Email collected (QID45, ForceResponse=ON) so the 6 engaged drop-offs are remindable; ?new=1 is the standing share-link rule; Study 2 sample = new=1 from 2026-06-26 onward.

📄 ops.py-driven

2026-06-29 milestone UPDATE 140 · UP cleaning re-done as v6; prereg v4.0 drafted (v3.4 never registered) ▸

Elad's personal code review of the v5-LOCKED UP cleaning pipeline caught real over-exclusion: the flat 60s duration rule wrongly DELETED 32 fully-completed responses, 23 of them M7/Q5 PRE rows from real in-study participants (staff_email + test_IP were clean). v5 (cleaning_2026-06-08) SUPERSEDED. New v6 plan (UP_QUALTRICS_ORGANIZATION_PLAN_v6_2026-06-29) is FINAL, awaiting Elad approval; on approval -> NEW readable from-scratch R code in a new folder (qualtRics readable export-tag names, no QID->rename; flag-not-delete <60s; drop IP filter entirely; staff=Elad/Shaked exact address only; no Yuval special-casing; drop responses before 2026-04-12; add skills_mX_all = mean of 7 items/module; anthropomorphism cite = Shen, Zhang, Fan & Liu 2024, NOT IDAQ). Value-level rename verified at data level (~28,500 cells, 0 mismatches). Prereg: v3.4 was NEVER registered on OSF; v4.0 drafted for the big Paper 2 only (within-subject + M-fam predictors + W-fam whole-study change + drop-M3-BIU robustness), sent to UP WhatsApp group, NOT registered. N expected up (mostly M7) once new code runs.

📄 ops.py-driven

2026-06-28 milestone UPDATE 139 · Clalit (Article 14): Zohar briefed in depth + forms sent ▸

Zohar briefed in depth via WhatsApp (2026-06-28) + sent נספח א' + the track-changes undertaking; open PI question (Elad/Karen/Zohar); awaiting Zohar+Eshkol consult before replying to Tziki. (Earlier 06-26 briefing deleted by Elad — wrong population/too shallow — corrected to parents-focused.)

📄 ops.py-driven

2026-06-26 milestone UPDATE 138 · Clalit consult + Yoav first Cesura run + RA flyer ▸

Three program updates (2026-06-26): (1) Article 14 Clalit - Elad confirms the IP / amended confidentiality undertaking needs a consult with BOTH Zohar AND Eshkol before replying to Tziki (was 'Zohar and possibly Eshkol'). (2) Article 5 Yoav - Danny Derby runs Yoav on CESURA for the FIRST time tomorrow, 2026-06-27, via his CPT channel = the Study 2 confirmatory cohort begins on Cesura (builds on the 2026-06-25 CalStudio->Cesura switch + verified sim link). (3) Program-level - a research-assistant (עוזר מחקר) recruitment flyer was published; existing task #52 refreshed to track applications/responses.

📄 ops.py-driven

2026-06-26 milestone UPDATE 137 · Article 4 bots LIVE on Elad's platform + new Yuval task set ▸

Article 4 bots (Carol 'carol bot -updated 260626' + Dafna V8) live on Elad's OWN platform (not Cesura); both got the 'End conversation' (סיים שיחה) button instruction + session_termination hard-stop; Dafna V7 = Carol-style conversational overhaul + gradual opening + side-window Gloria/Perls clip. New Yuval task set added: finish Article 10 by 2026-06-30 + meet (#67); organize Evelyn HE/EN questionnaires (#68); learn Prolific + the Evelyn run protocol (#69); build a GitHub capability (#70) + a questionnaire-building skill for Yuval (#71).

📄 ops.py-driven

2026-06-25 milestone UPDATE 136 · Article 4: socratic-training-bot skill built (TDD) + Dafna stance-bot V6 deployable ▸

NMS Training Agents (Carol/Jeff/Dafna). NEW reusable skill socratic-training-bot built via TDD (RED->GREEN, QA 11/11 PASS; gold example = Carol; composes simulation-architect METHOD + therapeutic-bot teaching feedback, NOT the patient-resistance engine, NOT Cesura). Dafna stances bot cleaned V3->V6: references stripped (slide#/cites/bibliography -> author-only footer), exact mandatory stage sequence, always-present Gloria/Perls clip link opened in a side window -- V6 deployable, Elad uploading to his OWN platform (mentiverse.ai, not Cesura). Maya/modes first-draft (Jeff slot) with Eshkol's NMS corrections pending Eshkol sign-off.

📄 ops.py-driven

2026-06-25 milestone UPDATE 135 · Yoav (Art 5): Cesura sim link verified live as participant ▸

Confirmed on the live survey (walked consent->signature->matrix->BOT page) that the BOT-block link now points to app.cesura.ai/sim/0c09ea55-... (text + href), no CalStudio remaining. The link had been mis-wired (visible text updated to Cesura but the href still pointed to pmfm.ai/CPTsimulator); Elad re-pointed it. Study 2 run on Cesura starts 2026-06-26.

📄 ops.py-driven

2026-06-25 milestone UPDATE 134 · Article 5 (Yoav): platform switch to Cesura; Study 2 run starts 2026-06-26 ▸

Yoav moves from CalStudio (pmfm.ai/CPTsimulator, performed poorly) to Cesura for the pre-registered Study 2 replication; new run starts 2026-06-26. Same questionnaire SV_039jMzxfdPX6yRU (9-item self-efficacy 1-10 pre/post); pre-reg splits the existing scale (Challenge subscale items 4/6/9 confirmatory). Cesura sim copy: app.cesura.ai/sim/0c09ea55-884e-4841-9459-5ca162a0c1cf. FLAG: survey BOT-block link TEXT shows the Cesura URL but the clickable href still points to CalStudio (pmfm.ai) - must swap href before the run. Study 2 confirmatory sample = Cesura cohort from 2026-06-26 onward, kept separate from CalStudio responses (Study 1 N=49; 354 auditable as of 2026-06-25).

📄 ops.py-driven

2026-06-25 milestone UPDATE 133 · Article 14 (Clalit) submission package assembled; undertaking IP clause revised in track-changes ▸

Per Tziki Shibak's 2026-06-23 email (Clalit legal counsel OK'd amending the undertaking wording), assembled one package 'לשליחה לכללית 2026-06-25/' (5 files): finalized protocol (clinical-manager placeholder removed, PI surname unified), filled נספח א' (לחתימה), and the כתב התחייבות סודיות with the IP clause revised IN-PLACE in track-changes (2 insertions, 0 deletions, author 'פרויקט Menti', 2026-06-25) so the יועמ"ש re-approves the wording. Carve-out excludes Menti's prior IP (bots/prompts/CPE/platform) + preserves academic publication. PI surname standardized to בנאקוט. Elad replies to Tziki himself; then Dalia signs -> district -> Helsinki. Admin/aggregate, no patient data. task #50.

📄 ops.py-driven

2026-06-25 milestone UPDATE 132 · Article 15 (Rina): non-course/control-group survey created + live; intervention at 29 responses / 13 completed from 06-18 run ▸

Created a SEPARATE non-course/control-group Qualtrics survey SV_3CQw2i9BhfgMuAm (MENTI - Rina Bot - Non-Course Version), PUBLISHED + ACTIVE, 0 responses, link .../jfe/form/SV_3CQw2i9BhfgMuAm. For control participants who did the Rina sim but were NOT in the course; SEPARATE SID/link from the intervention survey SV_734qjklvoJkl4to (NEVER pool). Recreated the live v29 definition MINUS the Course Evaluation block; QIDs re-mapped (consent QID26, college QID9). Inbal's 2026-06-23 answers applied + verified: Course Eval removed; Recommendation made bot-focused (0 'kurs' refs survey-wide); college options = Gordon/Shaanan/David Yellin/Other (controls from those colleges). Verified via participant browser walk + API deterministic re-reads. Word proposal then live link sent to Inbal (each once, signed Claude). Intervention survey response status (metadata only): auditable=29 (was 9 dev/test); 06-18 = 13 completed (Thursday pilot run); ~10 old dev/test responses (May 17-Jun 17) still in data and never wiped -> exclude at analysis by date/email.

📄 ops.py-driven

2026-06-21 milestone UPDATE 131 · Article 4 (Carol framework bot): DRAFT prepared + sent to Eshkol & Yuval ▸

A draft of the Carol bot (needs/Dweck framework, MentiTherapist triad) was prepared and sent to Eshkol and Yuval for feedback (per Elad 2026-06-21).

📄 ops.py-driven

2026-06-21 milestone UPDATE 130 · Article 10 body in-text citations verified (162/15/5) ▸

Verified every in-text citation in the manuscript BODY against its source: all 62 cited references obtained, one agent per reference, 182 citations checked -> 162 MATCH / 15 MISMATCH / 5 UNVERIFIABLE. 15 body mismatches include fabrications ([20] Levkovich '6%/94% sensitivity', [22] Weisman avatar example) and reversed claims ([10] Gonnermann-Muller, [30] Karnieli-Miller; [19] Elhilali wrong skill+models). Delivered v30 CONSOLIDATED docx (95 comments = 76 Table 1 + 19 body, 162 green) + Body_Citations_Findings to Yuval via WhatsApp once.

📄 ops.py-driven

2026-06-21 milestone UPDATE 129 · Article 10 Table 1 verified row-by-row + commented docx delivered to Yuval ▸

All 46 included papers in Table 1 (MANUSCRIPT_AdvSimulation_v28_TRACKED.docx) verified one agent per row: 46 cell agents (30 wrong cells, incl. 1 fabrication on row 15) + 46 Axis agents (~32 over-claims). Delivered v29 COMMENTED docx (76 comments, MENTI QC), tracked-changes alt, corrections ledger, verification report, and review_sources/ (46 PDFs). Sent commented docx to Yuval via WhatsApp once. Open: apply FIX comments, delete row 46 phantom, swap row 45 source PDF.

📄 ops.py-driven

2026-06-17 milestone UPDATE 128 · Article 15 (Rina) finalized to LIVE v29 + sent to Inbal; Article 14 Clalit district forms prepared ▸

Rina SV_734qjklvoJkl4to published v29 (Elad + Avi Assor): +Course Evaluation block (FIRST, 11-item, course_eval_7 reverse) + 4 required reflection items (QID30-33); de-counseled (strip 'ייעוץ'); Bot Quality 8->5, Learning 9->7, SUS 10->6 (no longer standard SUS, Elad accepted). All NON-DESTRUCTIVE (9 dev/test responses preserved). 21/21 deterministic API checks + participant-browser verified. New FINAL Word 2026-06-17 + codebook 0g. Sent to Inbal x2 (FINAL doc + survey link), signed (kalod). Open: QID19 'ייעוץ חינוכי' program option = Elad's decision. Clalit: 4 district forms filled (PI=Dalia Kelly Benacot) + IP-clause email drafted (Elad sends himself).

📄 ops.py-driven

2026-06-16 milestone UPDATE 127 · Session documentation pass 2026-06-15/16 ▸

Consolidated the 2026-06-15/16 session across 3 layers. Rina (A15) finalized to LIVE v27 (SUS+WAI added; identifier=mandatory email decided with Inbal #64 closed; v23 redesign merging Feedback+Perceived-Learning into 9-item learning block; Yuval's 2 reviewed edits applied v25; recommend_other removed v27); all dev/test data (9, zero real), wipe pending Elad's OK before Thu 2026-06-18 run (#6). Clalit (A14): Tziki Shibak's 2026-06-14 emails extracted - final pre-district round (Helsinki marking field + PI=Dalia Kelly Benacot + return with 4 forms); Elad does the corrections (#50). Dashboard edit-password prompt removed (auto-supply, commit dd13bfd). Morning briefing task #66 scheduled. Gmail-thread-reading lesson saved.

📄 ops.py-driven

2026-06-15 milestone UPDATE 126 · Article 15 (Rina): removed recommend_other (open-text) per Elad -> live v27 ▸

Removed recommend_other (QID25) from the Rina Recommendation block per Elad; moved to Trash non-destructively (data retained), kept the 3-row recommendation matrix (QID23). Live SV_734qjklvoJkl4to published v27; participant-browser verified; FINAL Word + codebook regenerated.

📄 ops.py-driven

2026-06-15 milestone UPDATE 125 · Article 15 (Rina): applied Yuval's 2 reviewed edits to live v25 ▸

Applied Yuval's 2 reviewed edits to Rina v24: consent IRB parenthetical removed + bot_quality_1 feminized. Live SV_734qjklvoJkl4to published v25; participant-browser verified verbatim; FINAL Word + codebook regenerated; 7 dev/test responses preserved.

📄 ops.py-driven

2026-06-15 milestone UPDATE 124 · 5-update catch-up (2026-06-15): Rina v23 + Cesura space, Narrative Review closing, Hili manuscript, Yoav→Cesura, Clalit email ▸

Processed 5 status updates from Elad. A15 Rina: questionnaire v23 done + with Yuval, sends tomorrow (gated on his review); new task #65 to prepare a Cesura Learning Space, final sim run Thu 2026-06-18 paired with questionnaire (= task #6 pilot). A10 Narrative Review: nearly complete, close + submit later this week (~2026-06-19; tasks #22/#23/#62 kept open). A17 Hili: Zohar works on the v3 manuscript this week. A5 Yoav: Yuval uploading the sim to Cesura (task #48 in progress). A14 Clalit: new email from Clalit received, proposal to be updated per it by Elad (task #50).

📄 ops.py-driven

2026-06-14 build UPDATE 123 · Article 15 (Rina) redesigned to live v23 ▸

Non-destructive redesign (0 data loss): merged Feedback+Perceived-Learning into 9-row 'תפיסת למידה ותועלת' (f2/f6/f7->learning_7/8/9; f1/3/4/5 dropped); reordered Consent->BotQuality->WAI->Learning->Recommendation->SUS->Open->Demog; educational register (dropped 'clinical'); time 15-20min. 7 dev/test responses preserved (not deleted). Browser-verified as participant; FINAL Word + codebook regenerated; Task #64 closed. AUDIT-PASSED pending Elad review.

📄 ops.py-driven

2026-06-14 milestone UPDATE 122 · Article 15 (Rina) v21: SUS + WAI-TECH-SF added; consent corrected ▸

Added SUS (10 items, 1-5) + WAI-TECH-SF (12 items, 1-7) to live Rina survey SV_734qjklvoJkl4to via additive API (no rebuild; all QIDs/data preserved). Fixed consent anonymity->confidential (email kept). Published v21. API + participant-browser (Playwright) verified. FINAL Word + codebook regenerated. Task #31 closed; #9 noted resolved for A15.

📄 ops.py-driven

2026-06-12 auto_refresh UPDATE 121 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-11 auto_refresh UPDATE 120 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-11 auto_refresh UPDATE 119 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-10 auto_refresh UPDATE 118 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-09 milestone UPDATE 117 · Article 17 (Hili) v3 manuscript AUDIT-PASSED + Word sent to Elad ▸

Submission-grade v3 of the parent-simulation pilot manuscript passed TWO independent multi-agent deep reviews (16 agents) + Crossref publisher-record verification: data forensically clean (every stat traces to stats.json), all 48 references real, all 9 of Elad's v3 emphases honored. 5 Crossref-verified citation fixes applied (Brown->Wu et al.; Chaby authors 7-8; Zeng 15 authors; Wang EMNLP DOI; Yirmiya et al. 2026 now PUBLISHED in Scientific Reports). Word built (TNR 12, double-spaced, 3 figures) + delivered to Elad in chat + via Telegram. STATUS = AUDIT-PASSED, awaiting Elad's USER-LOCK. Target: educational/school-psych Q1 Brief Report.

📄 ops.py-driven

2026-06-09 auto_refresh UPDATE 116 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-09 milestone UPDATE 115 · UP cleaning pipeline 7-clean LOCKED v5 (N=74) ▸

UP Qualtrics cleaning pipeline BUILT + RUN + 7-clean LOCKED (v5, git tag artifact-lock-v5-2026-06-08). N=74 clean participants (BIU 33 + Haifa 41), 669 LONG obs, WIDE 1 row/participant, FINAL ALL GREEN. 3 codebooks built (GUIDE/empirical/meaning-map, all PII-free). Analysis-ready for Paper 1 (12A, Shaked first, Cognitive Therapy and Research) + Paper 2 (12B, Elad first, npj Digital Medicine). Propagated from UP simu.

📄 ops.py-driven

2026-06-09 status_change UPDATE 114 · Correction to the 2026-06-08 meeting parse: 'Geffen Collins' = Yuval Holzman; Wed article = the Narrative Review (#10) ▸

Fixed owners on tasks #9/#56-#60 (Yuval/Geffen -> Yuval). Reframed the Wed 06-10 task to Article 10 (Narrative Review) finalize+submit with Zohar (article_ref repointed 0->10 via new ops.py task-update --article). Per Elad 2026-06-09.

📄 ops.py-driven

2026-06-08 milestone UPDATE 113 · Processed Elad+coordinator Research Sync (2026-06-08) into 7 new tasks + 8 updates ▸

Pulled the 104-min Timeless meeting; extracted action items across Articles 3/4/9/13/14/17 + Nature + Rina + lab ops. New tasks #56-#62; updated #4/#9/#10/#53/#55 + article next-steps for Evelyn(3)/Hili(17)/Clalit(14). NAME FLAG: 2nd speaker labeled 'Geffen Collins' = the Yuval coordinator role (confirm).

📄 ops.py-driven

2026-06-08 milestone UPDATE 112 · Dashboard: priority reorder + 3-layer propagation + drawer-scroll fix; Articles 9/13 tasks added ▸

Built drag/arrow task reordering (sort_order via /dashboard/order -> ops.py task-update --order). Dashboard edits now propagate to the linked article (layer 2) + a menti-agent drain log (layer 3). Fixed the drawer not scrolling to the bottom. Added ready-to-run tasks for Article 9 (CPE-vs-flat-prompt LIWC) and Article 13 (Prompt Compression). Backend rev 00006.

📄 ops.py-driven

2026-06-08 milestone UPDATE 111 · Task board organized: 5 new tasks + overdue flags + CLPsych dedup ▸

Added #48 Yoav->Cesura (Yuval, due 06-15), #49 Evelyn build (Yuval), #50 Clalit Barak-Gordon 7 comments (Elad+Yuval), #51 Study Studio review (Yuval), #52 Recruit RA (Elad). Closed #30 as superseded by #33 (duplicate CLPsych slides). Flagged 5 overdue (#14/#22/#23/#35/#33) with OVERDUE follow-up note + confirmed owners. this_week left as algorithm-derived (no manual-set subcommand exists; iron rule = no hand-edit).

📄 ops.py-driven

2026-06-08 auto_refresh UPDATE 110 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 milestone UPDATE 109 · Article 17 Hili pilot RAN + analyzed + preliminary report shared (N=29; both primaries supported) ▸

Pilot ran Sun 2026-06-07 ~12:00 IL @ MCLA Ramat Gan (MA Educational Psychology trainees; Lia parent-sim; within-subjects pre-post single arm). PRE 32 / POST 31 / linked N=29. H1 Scale C SE-for-guiding-parents up (d_z=0.55, p_holm=.013); H2 Scale B apprehension down (d_z=-0.47, p_holm=.018) - BOTH primaries supported (Holm). Challenge subscale d_z=.59 > Foundational d_z=.38 (Yoav-consistent). Acceptability composite 5.57/7; recommend 22/29=76% (p=.004). RFQ-8 flat (H7 null, uninformative). Password-protected preliminary report shared once to the parent-research group. All figures from Analysis_Pilot_2026-06-07/stats.json. Also: final clinical->therapeutic (קליני->טיפולי) cleanup on LIVE PRE v29/POST v25; Telegram scheduler #25/#32 Sunday spam root-caused (claim-after-run) + fixed (commit 706e311).

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 108 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 107 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 106 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 105 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 104 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-07 auto_refresh UPDATE 103 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-05 auto_refresh UPDATE 102 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-05 auto_refresh UPDATE 101 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-05 auto_refresh UPDATE 100 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-04 auto_refresh UPDATE 99 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-04 auto_refresh UPDATE 98 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-04 auto_refresh UPDATE 97 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-04 auto_refresh UPDATE 96 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-03 update UPDATE 95 · Yoav (Article 5): Cesura field-map Word doc built (v2, 6 fixes) ▸

Decomposed the Yoav CPT/Socratic prompt into Cesura's separate editor fields as a verbatim Word doc for manual paste-in. v2 adds 6 Elad-approved fixes incl. gender->pre-chat (שלב 0), opening->Starting Sentence, 10-interaction fix, and a NEW reflective Feedback Chat (inner world of Yoav, modeled on Lia). Verified: coverage + verbatim char-diff + 8-agent adversarial workflow + integrity (0 unauthorized changes). Delivered via Telegram + Dropbox (5.yoav/Prompt/).

📄 ops.py-driven

2026-06-03 update UPDATE 94 · Yoav (Article 5): Round 2 (Study 2) data check ▸

Live pull SV_039jMzxfdPX6yRU 2026-06-03: total raw 326. Round 1 (Study 1, pre-2026-05-25) raw 315 / finished 55 / clean N=49. Round 2 (Study 2, pre-reg 2026-05-25 onward) raw 11 / finished 4 / dur>=300 4. Only +4 valid completions since pre-reg vs N>=30 target; recruitment stalled.

📄 ops.py-driven

2026-06-03 auto_refresh UPDATE 93 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-03 auto_refresh UPDATE 92 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-02 update UPDATE 91 · Article 14 (Clalit): Barak Gordon review in; correction plan prepared ▸

Prof. Barak Gordon (Clalit Center District) returned 7 comments on the Clalit protocol (align to MoH AI guidelines; detail data-security incl. Cesura ISO-27001/no-training; model-training; chatbot error modes; risks; 'no research product' classification; name PI). Tzvika: PI must be Clalit employee+MA+GCP. Correction plan prepared (NOT applied) in 14.clalit/. Elad+Yuval discuss Wed 2026-06-03; plan sent to Yuval.

📄 ops.py-driven

2026-06-02 update UPDATE 90 · Article 17: landing page updated per Hili dry-run comments ▸

start.html: start CTA moved to bottom (read-all-first), gender-inclusive phrasing, clinical->therapeutic; pushed live. Elad adding 4-digit ID in Cesura; shared WhatsApp group opened (GATE A for Hili's Thursday comments).

📄 ops.py-driven

2026-06-02 milestone UPDATE 89 · Article 17 (Hili pilot) dry-run held 2026-06-02 — 8 decisions captured ▸

Dry-run Zoom (Elad+Yuval+Hili). LIVE Qualtrics PRE+POST + codebook + For_Hili + start.html + README verified internally consistent (Yuval SoT Scale C item 8 'מתאבל על'->'חווה מצוקה אל מול' + consent + Cesura 5de978bd all aligned; no drift). 8 decisions logged in _docs/PENDING_CHANGES_from_dryrun_2026-06-02.md: last-4-of-ID replaces email (Qualtrics+Cesura); קלינית->טיפולית; Scale C first-person; ADD curiosity measure pre+post (NEEDS ELAD DESIGN); demographics edits; pilot CONFIRMED Sun 2026-06-07 12:00; possible Developmental-Psych expansion. Wording changes flow OUTWARD from Yuval SoT docx AFTER Hili sends comments Thu 2026-06-04. Tasks #27/#28/#37 closed; 5 new tasks added.

📄 ops.py-driven

2026-06-02 session-summary UPDATE 88 · Article 17 Yuval edits + new Cesura URL applied 2026-06-02 ▸

Yuval Holzman uploaded Article_17_Questionnaire_02_0626 after Yuval.docx (09:42) — now SoT for Hili pilot. Track-changes XML extracted (DEL מתאבל+על / INS חווה מצוקה+אל מול). LIVE PRE+POST Scale C item 8 patched + republished. New Cesura URL (5de978bd-2831-4580-84a2-7f0c56f4b6de) replaces 3978eb32 in LIVE + 6 internal files + start.html. Site v15.21 live with Start tile restored. 2 lessons saved (cron-parser v5 API + track-changes XML reading).

📄 ops.py-driven

2026-06-02 auto_refresh UPDATE 87 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-02 auto_refresh UPDATE 86 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-02 auto_refresh UPDATE 85 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 84 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 83 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 82 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 81 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 80 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 79 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 78 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 77 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 76 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 75 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 74 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 73 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 72 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 71 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 70 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 69 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 68 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 67 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 66 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 65 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 64 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 63 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 62 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 61 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 60 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 59 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 58 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 57 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 56 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 55 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 54 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 53 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 52 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 51 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 50 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 49 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 48 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 47 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 46 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 45 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 44 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 43 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 42 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 41 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 40 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 39 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 38 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 37 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 36 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 35 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-06-01 auto_refresh UPDATE 34 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-05-31 session-summary UPDATE 33 · Session 2026-05-31 PM — multi-topic catch-up (CLPsych prep + Hili pilot wrap + MENTI website + budget + Evelyn) ▸

8 items: (1) CLPsych talk outline locked, Yuval to prep slides+script; (2) Demo Cesura sim built — Claude to record, copy for Hili; (3) MENTI website lacks contact form, needs fix; (4) Hili pilot setup done, Yuval to verify; (5) MENTI grant budget verified vs Israel Ministry of Science portal, updated PDF on Dropbox; (6) Evelyn budget tentative 120-150 NIS/participant pending actigraphy cost; (7) Email sent to Aya (CLPsych) re exact video upload deadline (~2026-06-04); (8) Research protocol sent to Hili.

📄 ops.py-driven

2026-05-30 status_change UPDATE 32 · Article 17 V2+RFQ-8 finalized, package delivered to Yuval ▸

25 sequential QC rounds with 5-consecutive-clean rule. Final streak achieved at round 25. Files delivered to Yuval Holzman via WhatsApp 2026-05-29 21:08 IL: For_Students_Introduction.docx, For_Hili_Article_Structure_Summary.docx, Article_17_Questionnaire_2026-05-28.docx, Article_17_Questionnaire_Items_2026-05-28.xlsx + Hebrew message. Site v15.12 live. Pilot Sun 2026-06-07; dry-run Tue 2026-06-02 13:00 Zoom. Blockers: Cesura URL (#34) + smoke test (#35).

📄 ops.py-driven

2026-05-28 milestone UPDATE 31 · Article 17 Qualtrics surveys BUILT ▸

Both surveys live on AI_PSYCH. PRE = SV_0xDcvoX33koSR14 (5 blocks, 18 questions). POST = SV_0ktHzcZDK8DVCe2 (6 blocks, 15 questions). Hebrew throughout, BIU look and feel, Hebrew RTL on every text element. PRE: consent + Scale C (9, 1-10) + Scale A (15, 1-6, Option D) + Scale B (6, 1-5 + open) + demographics + end-redirect with placeholder Cesura URL. POST: same scales + felt-SE (4, 1-7) + sim-feedback (7, 1-7) + 5 open-ended. EmbeddedData pre_response_id captured from URL. PENDING: Cesura URL to swap from PENDING-ARTICLE-17. Build scripts saved in folder + survey_ids.md + Hebrew drafts for Hili review at 2026-06-02 dry-run.

📄 ops.py-driven

2026-05-28 auto_refresh UPDATE 30 · Qualtrics auto-refresh: 9 updated, 1 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 9 surveys pulled fresh; 1 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-05-28 auto_refresh UPDATE 29 · Qualtrics auto-refresh: 4 updated, 6 blocked, 2 skipped ▸

ops.py qualtrics-refresh ran. 4 surveys pulled fresh; 6 returned 403/error (verify_needed=true); 2 TBD stubs ignored.

📄 ops.py-driven

2026-05-28 milestone UPDATE 28 · Yoav OSF live + Yuval CLPsych invited talk ▸

Yoav (Article 5) OSF pre-reg fz8xm is LIVE as of 2026-05-28 - Task #13 (approval confirmation) CLOSED, Task #12 reframed to active recruitment + weekly monitoring. New task added: Yuval Holzman prepares CLPsych 2026 invited-talk slides for Article 8 (Lia CPE methodology paper) by Sunday 2026-05-31.

📄 session_2026-05-28_deep_refresh.md

2026-05-28 cleanup UPDATE 27 · Task #21 (Omer OA) removed + Klalit -> Clalit spelling ▸

Elad reviewed open tasks: Omer has no Open Access affiliation arrangement so task is unnecessary - removed entirely from DB + dashboard. Also: standard Hebrew transliteration is Clalit (not Klalit) for the HMO; corrected throughout data.json.

📄 session_2026-05-28_deep_refresh.md

2026-05-28 milestone UPDATE 26 · Deep dashboard refresh + multiple article status updates ▸

Article 5 Yoav returned to active recruitment after OSF pre-reg fz8xm went live; Article 15 Rina SUS+WAI approved by Inbal (Yuval back-translation review is the gate before upload); MENTI sub-track chips/filter added to articles tab; folder-open CTA added to article drawers; Tasks #8 and #10 CLOSED, #9 reframed to Yuval back-translation review.

📄 session_2026-05-28_deep_refresh.md

2026-05-28 milestone UPDATE 25 · Article 17 protocol approved + Article 8 ACCEPTED + dashboard taxonomy deep-refresh ▸

Three big things: (1) Article 17 (Hili Pilot) protocol v5 APPROVED -- pilot 2026-06-08; (2) Article 8 (Lia CPE methodology) ACCEPTED at CLPsych -- Yuval Holzman invited for talk; (3) Article 10 (Narrative Review) in finalising stage. Dashboard deep-refreshed: Qualtrics counts pulled fresh (4 surveys from AI_PSYCH API; 6 surveys keep prior counts with verify_needed flag since they require workspace access), MENTI sub-track taxonomy (parent/therapists/foster/methodology) added to all articles, Article 12 split into 12A (Shaked-led pilot, id=121) + 12B (Elad-led NPJ DM, id=122). Tasks #14/15/19/20/21 repointed to 12B; #18 repointed to 12A (qualitative decision = Paper #1 scope).

📄 session_2026-05-28_dashboard_deep_refresh.md

2026-05-28 milestone UPDATE 24 · Article 17 (Hili Pilot) — protocol v5 APPROVED ▸

New MENTI study: MentiTherapist track. MA Educational Psychology trainees (N~40) play parent role in Lia bot, single 45-min session 2026-06-08 at Academic Center for Law & Business (Ramat Gan). Three scales pre/post: Scale C primary (SE for guiding parents, 9 items NEW, Yoav-style), Scale A indirect mechanism (Raviv-Bar-Tal 1995 Parental SE, Option D), Scale B obstacle (Kochavi apprehension 6+1). Plus post-only 4 felt-SE + 7 sim-feedback + 5 open-ended. IRB MENTI umbrella. Funding ~800 NIS Elad's lab. Build phase begins now (Qualtrics + Cesura copy of Lia). Dry-run 2026-06-02 13:00 Zoom. Article 16 stub also registered for count consistency (folder existed but unscoped).

📄 session_2026-05-28_article17_hili_protocol_approved.md

2026-05-27 milestone UPDATE 23 · Dashboard v2 LIVE — JSON-driven redesign + per-session SOP locked ▸

Phase 5 + 6 of dashboard redesign done. Swap to live at git commit f764bac. v1 archived as index_v1_backup_pre_swap.html. Article drawers now have [copy] + [open] file-path buttons. 4 tabs (Tasks/Articles/Pipeline/Activity), drawer-on-click, weekly Telegram audit Sundays 9am, 17 stale items fixed during extraction. Per-session method: edit data.json only; Stop hook auto-runs build + validate + push. Yuval Holzman notified via WhatsApp (DB-verified clean).

📄 session_2026-05-27_dashboard_redesign_full_method.md

2026-05-27 submission UPDATE 22 · Nature Awards 2026 SUBMITTED ▸

Nature Awards 2026 AI for Discovery application submitted via Nature portal (Health & Life Sciences). v4 frozen. verify_v4.py = 50 PASSES on 12 hard rules. Body-diff iron rule locked globally across 3 files.

📄 session_2026-05-27_nature_awards_SUBMITTED.md

2026-05-25 decision UPDATE 21 · UP Article 12 SPLIT INTO 2 PAPERS ▸

Shaked + Elad + Omer planning meeting decided to split UP into Paper #1 (pilot, Shaked-led) + Paper #2 (within-subject, Elad-led, NPJ DM HARD 2026-07-31). 8 new tasks #14-#21 added.

📄 session_2026-05-25_PM_UP_meeting_integration.md

2026-05-25 operations UPDATE 20 · UP tracking overhaul (N=71 canonical) + Joy session prep ▸

ID normalization SOT built (10 confirmed merges via IP cross-check). N=71 unique participants canonical (BIU=32, Haifa=39). Joy session prep delivered via WhatsApp (4 sends DB-confirmed).

📄 session_2026-05-25_yoav_prereg_submitted.md

2026-05-25 submission UPDATE 19 · Yoav Study 2 OSF pre-reg SUBMITTED ▸

Yoav Article 5 Study 2 AsPredicted pre-registration SUBMITTED to OSF by Elad. Registration fz8xm (DOI 10.17605/OSF.IO/fz8xm) pending OSF approval ~48h. Tasks #3 + #11 closed; #12 + #13 added.

📄 session_2026-05-25_yoav_prereg_submitted.md

2026-05-21 milestone UPDATE 18 · Yoav Study 2 pre-reg READY for submission ▸

Yoav Article 5 Study 2 AsPredicted pre-reg ready for Elad's pre-submission edits. Independent menti-domain review PASSED.

📄 session_2026-05-21_yoav_prereg_hooks_whatsapp.md

2026-05-20 analysis UPDATE 17 · Yoav v18 discovery-narrative Method+Results ▸

v18 Method+Results discovery-arc restructure with JSON-driven numbers. EFA at N=120. Iron rule: all numbers from code (v18_compute_all.R), not memory.

📄 session_2026-05-20_yoav_v18_discovery_narrative.md

2026-05-18 analysis UPDATE 16 · Yoav two-study pivot + v15 FINAL build ▸

Yoav Article 5 pivoted to two-study structure. Study 1 N=49 + Study 2 pre-reg. v15 Method+Results FINAL build with all stats computed in R.

📄 session_2026-05-18_yoav_two_study_pivot.md

2026-05-18 analysis UPDATE 15 · Yoav N=49 re-analysis + subscale FA + EFA at N=120 ▸

Fresh re-analysis at N=49. Primary moved away from significance (t=1.432, p=.159). Challenge subscale survives (d=0.38, p=.010). EFA at N=120: 1-factor stable.

📄 session_2026-05-18_yoav_reanalysis_N49.md

2026-05-18 operations UPDATE 14 · Article 15 Rina v19 — Article# removed + SUS/WAI proposal to Inbal ▸

Rina v19 published: removed 'Article 15' from participant-facing strings + SUS/WAI proposal sent to Inbal. 7/8 Inbal edits applied via v18. Tasks #8/#9/#10 awaiting Inbal replies.

📄 session_2026-05-17_article15_rina_build.md

2026-05-17 operations UPDATE 13 · Article 15 Rina BUILT end-to-end ▸

Article 15 (Rina counseling-student bot, SDT) built end-to-end. Qualtrics SV_734qjklvoJkl4to live + active. v8 published (82/82 verification checks).

📄 session_2026-05-17_article15_rina_build.md

2026-05-17 operations UPDATE 12 · Three coordinated tasks: Yuval EN letter + Carol/Jeff/Dafna task + Hili Gmail draft ▸

WhatsApp EN letter to Yuval (DB-verified clean). Carol/Jeff/Dafna status-check registered as Task #4. Gmail draft to Hili created (draft ID r9100630670592223174).

📄 session_2026-05-17_three_tasks_hili_followup.md

2026-05-17 operations UPDATE 11 · UP pre-reg v3.4 sent to סימולציות UP group ▸

UP pre-registration v3.4 incorporating all 27 Shaked review comments sent to UP team WhatsApp group with Hebrew change-summary. DB-verified clean.

📄 session_2026-05-17_up_prereg_v34.md

2026-05-17 operations UPDATE 10 · Outreach EN translation for Hili ▸

English translation of Elad's post-edit Hebrew letter for Hili Kochavi (MCLA, prospective UP site 3).

📄 session_2026-05-17_outreach_letter_EN.md

2026-05-17 operations UPDATE 9 · UP pre-reg v3.3 (signature violation -> rule) ▸

UP pre-reg v3.3 build had signature violation that triggered the always-sign-as-Claude permanent rule.

📄 session_2026-05-17_up_prereg_v33.md

2026-05-15 analysis UPDATE 8 · Post-vacation Part 2 follow-up: CLP reverted to Under Review + UP deep audit ▸

CLPsych Article 8 reverted from 'review passed' (inferred) to 'Under Review' (Elad correction). UP true clean N=363 via deep audit. Lavidor IRB covers Articles 3/7/9/13.

📄 session_2026-05-15_part2_followup.md

2026-05-15 operations UPDATE 7 · Post-vacation status review ▸

Post-vacation status review (data state freeze for 2026-05-15) — UP S3+S4 sprint complete; Evelyn IRB cleared (Israel-only, no UCL needed).

📄 session_2026-05-15_status_review.md

2026-05-27 infrastructure UPDATE 6 · Post-submission dashboard reorg planning ▸

Audit + redesign planning after Nature Awards submission. data.json source-of-truth file scoped; phase 1 extraction in progress.

📄 session_2026-05-27_post_submission_dashboard_reorg.md

2026-05-18 milestone UPDATE 5 · Article 15 Rina v18 Inbal edits applied ▸

Inbal Cohen returned v15 questionnaire with 9 inline red-text annotations. 7 edits applied directly, 1 modified, 1 flagged as Task #8 (Bot Quality item 2 reword). v18 published.

📄 session_2026-05-17_article15_rina_build.md

2026-05-18 milestone UPDATE 4 · Article 15 Rina v12 restructure (Perceived Learning) ▸

Per Elad's Option A: Removed Self-Efficacy block (9 items), Added Perceived Learning block (6 items), Expanded Bot Quality 4 -> 8 items. v15 published.

📄 session_2026-05-17_article15_rina_build.md

2026-05-17 milestone UPDATE 3 · Article 15 Rina gender-inclusive Hebrew + Look-and-Feel + 4-fix ▸

40 feminine-only phrases -> gender-inclusive slash forms. BIU 4-logo banner removed. Open-ended 1-3 made mandatory; recommendation converted to Matrix/Likert; ריינה -> רינה.

📄 session_2026-05-17_article15_rina_build.md

2026-05-17 milestone UPDATE 2 · Article 15 Rina consent fix ▸

Removed em-dashes + smart quotes from welcome + consent; promoted BIU IRB approval sentence to first paragraph. v6 published. Task #5 (IRB scope) closed.

📄 session_2026-05-17_article15_rina_build.md

2026-05-17 milestone UPDATE 1 · Article 15 (Rina) initial build ▸

New article on Rina counseling-student bot evaluation (SDT/needs-supportive dialogue). Qualtrics SV_734qjklvoJkl4to live. 82/82 verification checks pass.

📄 session_2026-05-17_article15_rina_build.md

2026-07-16T01:40:09 | 1 (0 block, 1 warn)

Each finding shows a preferred ops.py fix command + alternatives. Click "Copy" to copy the command to your clipboard; run it in a terminal in menti-dashboard/. No "Apply" button — every state change must go through ops.py manually (iron rule).

🟡 WARN staleness #3969fb31d4b4

article id=17 stale 31d (last_touched=2026-06-15) -- WARN

⭐ Commit a real next step

python ops.py article-update --id 17 --next-step "<one-line next step>"

Task #1 · CLOSED

Reminder: Mital and Avi

Status CLOSED 2026-03-10

סמן

עדכון תאריך

הערות

Owner Elad

Type one_time / one_time

Last touched 2026-03-10

Description

Legacy reminder.

History

2026-03-10: fired and disabled (one_time)

Source

telegram-bot.db · created 2026-03-10 14:32:33

Task #2 · CLOSED

Reminder: Mital and Avi 22:32

Status CLOSED 2026-03-10

סמן

עדכון תאריך

הערות

Owner Elad

Type one_time / one_time

Last touched 2026-03-10

Description

Legacy reminder.

History

2026-03-10: fired and disabled (one_time)

Source

telegram-bot.db · created 2026-03-10 20:27:50

Task #3 · CLOSED

Yoav Article 5 publication path - Study 1 draft + Study 2 pre-reg SUBMITTED OSF fz8xm; paper PAUSED until next CPT cohort arrives ~2026-05-25

Status CLOSED 2026-05-25

סמן

עדכון תאריך

הערות

Owner Elad

Type decision_flag / decision_flag

Article #5 — Socratic Yoav

Last touched 2026-05-15

Description

Yoav Article 5 publication path. Closed 2026-05-25: Study 1 draft READY + Study 2 OSF pre-reg SUBMITTED (fz8xm); paper PAUSED until next CPT cohort arrives.

Source

telegram-bot.db · created 2026-05-15 15:15:05

Task #4 · CLOSED

Carol / Jeff / Dafna framework bots — status check

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Yuval (chase)

Deadline 2026-06-09 (12 days)

Type status_check / status_check

Article #4 — Reimagining Clinical Training (Carol+Dweck triad)

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-02

Description

Carol / Jeff / Dafna framework bots — status check. Yuval asked 2026-05-17 to update this week. Per audit row 36: VERIFY NEEDED (10+ days stale).

History

2026-07-02: CLOSED — Resolved: Carol (260626) + Dafna V8 live on Elad's platform 2026-06-25/26; Maya/modes draft awaits Eshkol sign-off (tracked on Article 4)
2026-06-08: Meeting 2026-06-08: coordinator to meet Dafna + revise the prompt per Eshkol's workshop presentation; aim to finish the Carol/Dafna bot soon.
2026-06-08: deadline -> 2026-06-09 (by אלעד)
2026-06-08: 2026-06-08: owner confirmed Yuval (Elad to chase). Still open - status check pending.

Source

telegram-bot.db · created 2026-05-17 10:08:42

Task #5 · CLOSED

Article 15 Rina - IRB scope (covered under MENTI umbrella)

Status CLOSED 2026-05-17 PM

סמן

עדכון תאריך

הערות

Owner Elad

Type decision_flag / decision_flag

Article #15 — Rina

Last touched 2026-05-17

Description

Article 15 Rina IRB scope confirmation. Closed 2026-05-17 PM (covered under MENTI umbrella per Elad).

Source

telegram-bot.db · created 2026-05-17 12:30:36

Task #6 · CLOSED

Article 15 Rina - Pilot test with 5-10 counseling students before full cohort launch

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Inbal + Elad

Deadline 2026-06-18 (21 days)

Type task / one_time

Article #15 — Rina

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-02

Description

Article 15 Rina — pilot test with 5-10 counseling students before full cohort launch. IRB now cleared (was the upstream blocker). Per audit row 32: VERIFY NEEDED (no recruitment movement reported).

History

2026-07-02: CLOSED — Rina first real run 2026-06-18 (13 completed); collection continuing on live v29 - pilot gate passed
2026-06-08: deadline -> 2026-06-14 (by אלעד)

Source

telegram-bot.db · created 2026-05-17 12:30:36

Task #7 · CLOSED

Article 15 Rina - Assor identity CONFIRMED: Prof. Avi Assor (Ben-Gurion University, 054-577-3966); Inbal Cohen is field coordinator at David Yellin (054-795-3264)

Status CLOSED 2026-05-18

סמן

עדכון תאריך

הערות

Owner Elad

Type lookup / lookup

Article #15 — Rina

Last touched 2026-05-17

Description

Article 15 Rina Assor identity. Closed 2026-05-18: Prof. Avi Assor (Ben-Gurion U, 054-577-3966); Inbal Cohen field coordinator at David Yellin (054-795-3264).

Source

telegram-bot.db · created 2026-05-17 12:30:36

Task #8 · CLOSED

Article 15 Rina - Bot Quality item 2 (proposal sent to Inbal 2026-05-18, awaiting reply)

Status CLOSED 2026-05-28

סמן

עדכון תאריך

הערות

Owner Inbal

Type decision_flag / decision_flag

Article #15 — Rina

Verify ⚠ Verification needed — info may be stale

Last touched 2026-05-28

Description

Article 15 Rina — Bot Quality item 2 reword proposal sent to Inbal 2026-05-18, awaiting reply (9 days stale).

History

2026-05-28: CLOSED - Inbal approved both questionnaires (SUS + WAI-TECH-SF)

Source

telegram-bot.db · created 2026-05-19 07:10:20

Task #9 · CLOSED

Article 15+ Rina: SUS Hebrew translation decision - review back-trans + 4 flagged items + decide adoption for MENTI portfolio

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Yuval

Deadline 2026-06-11 (14 days)

Type decision_flag / decision_flag

Article #15 — Rina

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-02

Description

Inbal approved SUS adoption 2026-05-28. Before the SUS block is uploaded to the live survey, Yuval reviews the Hebrew back-translation (already drafted; needs her sign-off).

History

2026-07-02: CLOSED — Resolved by events: SUS adopted into Article 15 live survey (v27; trimmed to 6-item variant in v29 with Elad's acceptance)
2026-06-14: 2026-06-14: Yuval's finalized SUS Hebrew wording now LIVE in Rina survey v21 (item 5 kept as 'hamaarechet/system' per Yuval). Effectively resolved for Article 15. Portfolio-wide SUS-Hebrew adoption decision (Yoav/Evelyn/Lia/UP) still Elad's call.
2026-06-08: Meeting 2026-06-08: coordinator does the 10-item SUS back-translation, to Elad for sign-off by Thu 2026-06-11.
2026-06-08: "C:\Users\user\Dropbox\ARLAB shared folder\PhD students\EladR\menti\study\articles\15. Rina\Inbal_edits\SUS_Hebrew_back_translation_review_2026-05-18.xlsx" ליובל לבדוק (note by אלעד)
2026-06-08: deadline -> 2026-06-11 (by אלעד)
2026-06-08: 2026-06-08: owner confirmed Yuval (back-translation review gate before SUS upload).
2026-05-28: Inbal approved; reframed task to Yuval back-translation review (gate before upload)

Source

telegram-bot.db · created 2026-05-19 08:09:46

Task #10 · CLOSED

Article 15 Rina: validated-Q proposal decision - SUS yes/no + WAI-TECH-SF yes/no (sent to Inbal 2026-05-18)

Status CLOSED 2026-05-28

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-14 (17 days)

Type decision_flag / decision_flag

Article #15 — Rina

Verify ⚠ Verification needed — info may be stale

Last touched 2026-06-08

Description

Article 15 Rina — validated-Q proposal decision: SUS yes/no + WAI-TECH-SF yes/no. Sent to Inbal 2026-05-18. 9 days stale.

History

2026-06-08: Meeting 2026-06-08: Elad to close questionnaire approval with Inbal by 2026-06-14.
2026-05-28: CLOSED - Inbal approved both SUS + WAI joint adoption; remaining work is upload (handled by new task)

Source

telegram-bot.db · created 2026-05-19 08:09:46

Task #11 · CLOSED

Yoav Article 5 Study 2 pre-reg SUBMITTED to OSF (fz8xm); DOI 10.17605/OSF.IO/fz8xm; pending approval ~48hr

Status CLOSED 2026-05-25

סמן

עדכון תאריך

הערות

Owner Elad

Type action_required / one_time

Article #5 — Socratic Yoav

Last touched 2026-05-19

Description

Yoav Article 5 Study 2 pre-reg SUBMITTED to OSF (fz8xm); DOI 10.17605/OSF.IO/fz8xm. Closed 2026-05-25 (Elad clicked Register).

Source

telegram-bot.db · created 2026-05-19 20:11:56

Task #12

Yoav Article 5 - Active recruitment + weekly monitoring (OSF fz8xm live 2026-05-28)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type action_required / one_time

Article #5 — Socratic Yoav

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-02

Description

OSF pre-reg fz8xm is live as of 2026-05-28. Yoav back to active recruitment via CPT training courses. Monitor weekly response counts on SV_039jMzxfdPX6yRU (now at 317, up from 312 on 2026-05-18). Target N>=30 confirmatory; current Study 2 cohort just starting.

History

2026-07-02: 2026-07-02 (Elad): instead of a reminder email to the 5 engaged partials - the team will hold another live run/meeting (session) for Yoav data collection. Reminder email dropped for now; date TBD
2026-07-02: 2026-07-02: Cesura era - Friday 06-26 run: 20 started / 8 completed / 12 partial (all completers new=1); reminder email to 6 engaged partials awaits Elad's go; ?new=1 link standing rule
2026-05-28: Yoav RETURNED TO RUNNING - OSF live, recruitment restart, monitor responses weekly

Source

telegram-bot.db · created 2026-05-25

Task #13 · CLOSED

Yoav Article 5 Study 2 - Confirm OSF registration fz8xm went PUBLIC after 48hr approval window (deadline ~2026-05-27)

Status CLOSED 2026-05-28

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-05-27 (1 days overdue)

Type lookup / lookup

Article #5 — Socratic Yoav

Verify ⚠ Verification needed — info may be stale

Last touched 2026-05-28

Description

Yoav Study 2 — confirm OSF registration fz8xm went PUBLIC after 48h approval window. Visit https://osf.io/fz8xm/ — should resolve to public, immutable record with DOI 10.17605/OSF.IO/fz8xm active.

History

2026-05-28: CLOSED - OSF fz8xm went live and active, registration confirmed public

Source

telegram-bot.db · created 2026-05-25

Task #14 · CLOSED

UP Article 12 - Joy (Haifa) data completion (precondition for analyses)

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-05-30 (2 days)

Type action_required / one_time

Article #122 — UP Within-Subjects (Paper #2)

Verify ⚠ Verification needed — info may be stale

Last touched 2026-06-09

Description

UP Article 12 — Joy (Haifa) data completion (precondition for Paper #1 + Paper #2 analyses). 39 enrolled / 12 done / 27 need follow-up. Joy session prep delivered via WhatsApp 2026-05-25 evening (4 sends DB-confirmed). Joy in-person Tue 2026-05-26.

History

2026-06-09: 2026-06-09: UP cleaning pipeline 7-clean LOCKED v5 ran on available data (N=74; v5 recovered 2 >=90% partials). Task already CLOSED by Elad 2026-06-08; left closed. Note: 3 participants retained without consented Q-Pre row (BIU 7813; Haifa 8287, 6465) flagged for analysis stage.
2026-06-08: CLOSED — marked done from dashboard by אלעד
2026-06-08: 2026-06-08: OVERDUE - follow up. Owner confirmed: Elad.

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #15 · CLOSED

UP codebook - Q6 missing one question; sync codebook with v3.4 pre-reg before primary analyses

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-09 (12 days)

Type task / one_time

Article #122 — UP Within-Subjects (Paper #2)

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-02

Description

UP codebook — Q6 missing one question; sync codebook with v3.4 pre-reg before primary analyses. Quick fix.

History

2026-07-02: CLOSED — SUPERSEDED by the v6 cleaning rebuild. This task's premise (sync the codebook with the v3.4 pre-reg) is obsolete: (a) prereg v3.4 was NEVER registered on OSF and has been replaced by the v4.0 draft; (b) the entire cleaning pipeline was re-done from scratch as v6 (cleaning_v6_2026-06-30/, BUILT + AUDIT-PASSED 7/7 on 2026-07-02) with its OWN generated clean codebook (data/clean/codebook_clean) built from readable qualtRics export-tag names. Any codebook/Q6 alignment now happens against v6 + prereg v4.0, not v3.4. Evidence: UP docs/SESSION_HANDOFF_2026-07-02.md + session_2026-06-29_up_v6_recleaning_prereg_v4.md.
2026-06-09: 2026-06-09: UP cleaning lock v5 built 3 data-derived codebooks (GUIDE/empirical/meaning-map) - these describe the cleaned data, NOT the pre-reg. The Q6-missing-question vs v3.4 pre-reg reconciliation is NOT shown resolved in the lock; LEFT OPEN. Use codebook_data_derived_items.csv to check Q6 item coverage against the pre-reg.
2026-06-08: deadline -> 2026-06-09 (by אלעד)

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #16 · CLOSED

UP - send Shaked + Omer the MENTI paper (Yirmiya 2026) + Tal paper (Elyoseph foster care, under review) for reference

Status CLOSED 2026-05-31

סמן

עדכון תאריך

הערות

Owner Elad

Type task / one_time

Article #122 — UP Within-Subjects (Paper #2)

Last touched 2026-05-31

Description

UP — send Shaked + Omer the MENTI paper (Yirmiya 2026 Scientific Reports) + Tal foster-care paper (SSRN 6207845). Closed 2026-05-25 PM: Elad sent both himself via WhatsApp 15:46-15:47 same day.

History

2026-05-31: CLOSED — Closed 2026-05-25 PM by Elad himself; he sent the WhatsApp messages directly. JSON status sync from DB 2026-05-31.

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #17 · CLOSED

UP - send Shaked + Omer the CFP links (NPJ Digital Medicine + others) via WhatsApp

Status CLOSED 2026-05-31

סמן

עדכון תאריך

הערות

Owner Elad

Type task / one_time

Article #122 — UP Within-Subjects (Paper #2)

Last touched 2026-05-31

Description

UP — send Shaked + Omer the CFP links (NPJ Digital Medicine + 2 other Nature Collections). Closed 2026-05-25 PM: Elad sent 5 CFP messages 16:00-16:03 + 21:30-21:32 same day.

History

2026-05-31: CLOSED — Closed 2026-05-25 PM by Elad himself; he sent the CFP links via WhatsApp. JSON status sync from DB 2026-05-31.

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #18 · CLOSED

UP - Tal input on whether to add qualitative analysis (open-response themes) to Paper #1

Status CLOSED 2026-07-06

סמן

עדכון תאריך

הערות

Owner Elad (chase Tal)

Type decision_flag / decision_flag

Article #121 — UP Pilot (Paper #1)

Verify ⚠ Verification needed — info may be stale

Last touched 2026-07-06

Description

Paper #1 (M3 single-sim pilot + qualitative; Shaked first author; Cog. Therapy & Research; NOT pre-registered). The 2026-06-29 Elad+Shaked meeting confirmed Tal Eliyosef LEADS the qualitative (open-response themes) arm for Paper #1. Still pending Tal's actual input/coding. STILL VALID.

History

2026-07-06: CLOSED — Premise removed 2026-07-02: the M3 single-sim pilot (12A) that this qualitative arm attached to is DROPPED. Per the 2026-07-02 call the course data is analyzed as ONE big paper (12B); qualitative (Tal Eliyosef) folds into the big-paper story if used at all -- no separate Paper #1 to scope it into.
2026-07-02: 2026-07-02: confirmed still valid + clarified as Paper 1's qualitative arm (Tal Eliyosef leads, per 06-29 meeting) during UP task reconciliation.
2026-06-09: 2026-06-09: Cleaning data now LOCKED + analysis-ready (N=74). Qualitative add-on for Paper #1 (12A) still pending Tal Elyoseph's input (8.6 meeting). Decision still OPEN - unblocks Paper #1 scope lock.

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #19 · CLOSED

UP - Next team meeting 2026-06-08 14:00 - review Paper #1 (pilot) results

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-09 (12 days)

Type action_required / one_time

Article #122 — UP Within-Subjects (Paper #2)

Last touched 2026-07-02

Description

UP — next team meeting 2026-06-08 14:00 (Shaked + Elad + Omer). Agenda: review Paper #1 (Shaked-led pilot) preliminary results.

History

2026-07-02: CLOSED — Date passed and superseded. The 2026-06-08 review meeting is moot; the paper-structure / cleaning-strategy review was carried by a later Elad+Shaked meeting on 2026-06-29 ('AI Bot Research Paper Structure', pulled from Timeless, COMPLETED), which confirmed the v6 cleaning decisions + the two-paper / robustness-test / pre-registration strategy. The remaining forward-looking meeting is a 3-way meeting WITH OMER (paper structure + prereg + asking Omer to drop the 28-July deadline), still to be scheduled by Shaked -- that is a distinct new item, flagged for Elad rather than kept under this stale row. Evidence: UP docs/SESSION_HANDOFF_2026-07-02.md sec E + OPEN ITEMS.
2026-06-08: deadline -> 2026-06-09 (by אלעד)

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #20 · CLOSED

UP - Submit Paper #2 (within-subject, large) to NPJ Digital Medicine - HARD deadline 2026-07-31

Status CLOSED 2026-07-06

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-07-31 (64 days)

Type action_required / hard_deadline

Article #122 — UP Within-Subjects (Paper #2)

Blockers #14 #15 #18

Last touched 2026-07-06

Description

Paper #2 = UP within-subject (WITH vs WITHOUT simulation x modules x persistent Ethan) + predictors/moderators + whole-study change; Elad first author; target npj Digital Medicine special issue. Pipeline state (2026-07-02): v6 cleaning BUILT + AUDIT-PASSED 7/7 (N=76 = BIU 34 + Haifa 42, 702 LONG obs) but NOT yet USER-LOCKED (awaiting Elad's personal review). Pre-reg = v4.0 DRAFTED (preregistration_draft_v4.0) but NOT registered on OSF; after USER-LOCK, update prereg v4.0 sec6 exclusions + N to 76/702. HARD deadline 2026-07-31.

History

2026-07-06: CLOSED — npj 31-Jul-2026 HARD deadline DROPPED in the 2026-07-02 Elad+Omer+Shaked call (all agreed unattainable). Big paper (12B) target switched to JMIR Medical Education with NO hard deadline; Nature Mental Health option left open. Superseded by new task for the JMIR Med Ed submission.
2026-07-02: 2026-07-02: refreshed to v6/prereg-v4 reality (N=76/702 AUDIT-PASSED not user-locked; prereg v4.0 drafted not registered) + set due 2026-07-31 during UP task reconciliation.
2026-06-09: npj Digital Medicine deadline CONFIRMED = 2026-07-31 per the CFP site (Elad verified 2026-06-09). Supersedes the UP lock file's '24 July' and Shaked's email '26 June'. No change (dashboard already 07-31).

Source

telegram-bot.db · created 2026-05-25 23:00:00

Task #22 · CLOSED

Yuval narrative review (Article 10) - she received comments and is preparing v19 for Sunday review

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Yuval Holzman

Deadline 2026-06-10 (13 days)

Type status_check / status_check

Article #10 — Narrative Review (5-Axis)

Last touched 2026-07-02

Description

Yuval narrative review (Article 10) -- v19 ready after incorporating reviewer comments. Final stages; submission imminent after Sunday close-out meeting.

History

2026-07-02: CLOSED — Superseded: v30 consolidated (95 comments) delivered to Yuval 2026-06-21; closing tracked by task 67
2026-06-08: deadline -> 2026-06-10 (by אלעד)
2026-06-08: 2026-06-08: OVERDUE - follow up. Owner confirmed: Yuval Holzman.

Source

telegram-bot.db · created 2026-05-25 19:23:04

Task #23 · CLOSED

Yuval Article 10 review meeting - Sunday review of v19 to close narrative review

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad + Yuval

Deadline 2026-06-10 (13 days)

Type action_required / one_time

Article #10 — Narrative Review (5-Axis)

Last touched 2026-07-02

Description

Yuval Article 10 review meeting Sunday 2026-05-31 -- walk through v19 of narrative review on LLM-simulated patients. Goal: close QA loop and finalise for submission.

History

2026-07-02: CLOSED — Superseded by task 67 (finish Article 10 by 2026-06-30 then Elad+Yuval meet)
2026-06-08: deadline -> 2026-06-10 (by אלעד)
2026-06-08: 2026-06-08: OVERDUE - follow up. Owner confirmed: Elad + Yuval.

Source

telegram-bot.db · created 2026-05-25 19:23:04

Task #24

Nature Awards 2026 outcome - track decision (SUBMITTED via Nature portal 2026-05-27; notification expected per Nature programme timeline; owner Elad)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type status_check / status_check

Last touched 2026-07-02

Description

Nature Awards 2026 outcome — SUBMITTED 2026-05-27 via Nature portal (Health & Life Sciences category). Track decision (notification per Nature programme timeline, typically weeks to months). Frozen artifact: NatureAwards_Application_v4.docx.

History

2026-07-02: 2026-07-02: still awaiting Nature decision (nothing received)

Source

telegram-bot.db · created 2026-05-27 09:30:00

Task #26 · CLOSED

Article 17 Hili pilot - Build pre + post Qualtrics surveys on AI_PSYCH (3 scales + post-only blocks + ResponseId redirect chain)

Status CLOSED 2026-05-30

סמן

עדכון תאריך

הערות

Owner Claude / menti-agent

Deadline 2026-05-31 (3 days)

Type task / one_time

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-05-30

Description

Surveys BUILT 2026-05-28 on AI_PSYCH. PRE=SV_0xDcvoX33koSR14 (18 Qs), POST=SV_0ktHzcZDK8DVCe2 (15 Qs). PENDING: Cesura URL from Elad to swap from PENDING-ARTICLE-17 placeholder + Hili review of post-block Hebrew at dry-run.

History

2026-05-30: CLOSED — V2+RFQ-8 final package delivered to Yuval 2026-05-29 21:08 IL. Both LIVE surveys final (PRE 5 blocks 17Q, POST 6 blocks 14Q, RFQ-8 added, Scale A dropped, Scale B 1-7, sim_feedback_3 to המדריכה). Files clean of forbidden tokens. 25 QC rounds + 5-clean streak achieved.

Source

telegram-bot.db · created 2026-05-28 15:00:00

Task #27 · CLOSED

Article 17 Hili pilot - Cesura copy of Lia bot for analytics isolation (owner: Elad, due 2026-06-01)

Status CLOSED 2026-06-02

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-01 (4 days)

Type action_required / one_time

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-02

Description

Article 17 (Hili pilot) — Elad creates a SEPARATE Cesura copy of the Lia bot for Article 17 cohort (analytics + transcripts stay distinct from any other Lia deployments). New sim ID to be assigned; prompt unchanged. Elad sends the new sim URL to Claude for wiring into the three-step ribbon (pre Qualtrics → Cesura → post Qualtrics with ResponseId).

History

2026-06-02: CLOSED — Cesura copy of Lia created; sim 5de978bd-2831-4580-84a2-7f0c56f4b6de wired into LIVE PRE EOSRedirect + start.html + README + codebook. NOTE: still needs last-4-of-ID collection added Cesura-side per dry-run decision #1 (tracked in new task).

Source

telegram-bot.db · created 2026-05-28 15:00:00

Task #28 · CLOSED

Article 17 Hili pilot - Dry-run meeting (Elad + Hili + Yuval) Zoom 2026-06-02 13:00

Status CLOSED 2026-06-02

סמן

עדכון תאריך

הערות

Owner Elad + Hili + Yuval

Deadline 2026-06-02 (5 days)

Type action_required / one_time

Article #17 — Hili Pilot — MentiTherapist

Blockers #26 #27

Last touched 2026-06-02

Description

Article 17 (Hili pilot) — dry-run meeting Tue 2026-06-02 13:00 Zoom. Hili reviews Scale C v1 items. Elad walks the full pre → Cesura → post flow as a participant. Issues raised → fix list before 2026-06-04. Blocked on Tasks #26 (Qualtrics build) and #27 (Cesura copy).

History

2026-06-02: CLOSED — Dry-run held 2026-06-02 13:00 Zoom (Elad+Yuval+Hili). Materials reviewed; 8 decisions captured in _docs/PENDING_CHANGES_from_dryrun_2026-06-02.md

Source

telegram-bot.db · created 2026-05-28 15:00:00

Task #29 · CLOSED

Article 17 Hili pilot - Run pilot session in Hili's class Sun 2026-06-07 (final class of Semester B; ~40 students; 45-min slot)

Status CLOSED 2026-06-07

סמן

עדכון תאריך

הערות

Owner Elad + Hili + Yuval

Deadline 2026-06-07 (10 days)

Type action_required / one_time

Article #17 — Hili Pilot — MentiTherapist

Blockers #26 #27 #28

Last touched 2026-06-07

Description

Article 17 (Hili pilot) - run the pilot session in Hili's final class of Semester B. 45-min slot at MCLA Ramat Gan. ~40 MA Educational Psychology students. PRE Qualtrics (5 blocks, 17 Qs, ~10 min) + Cesura sim with Lia (~25 min) + POST Qualtrics (6 blocks, 14 Qs, ~10 min). Blocked on #27 (Cesura URL) + #35 (smoke test).

History

2026-06-07: CLOSED — Pilot RAN Sun 2026-06-07 ~12:00 IL @ MCLA Ramat Gan; N=29 linked; analyzed same day; both primaries supported; report shared

Source

telegram-bot.db · created 2026-05-28 15:00:00

Task #31 · CLOSED

Upload SUS + WAI-TECH-SF blocks to Article 15 Rina live survey

Status CLOSED 2026-06-14

סמן

עדכון תאריך

הערות

Owner Claude/menti-agent

Deadline 2026-06-10 (13 days)

Type / build

Article #15 — Rina

Blockers #9

Last touched 2026-06-14

Description

After Yuval signs off on the SUS Hebrew back-translation (Task #9), build SUS (10 items, 1-5, reverse-coded 2/4/6/8/10) + WAI-TECH-SF (12 items, 1-7, Bond/Goal/Task subscales) blocks into live Rina survey SV_734qjklvoJkl4to. Same SurveyID, no participant disruption.

History

2026-06-14: CLOSED — Done 2026-06-14: SUS (10 items, 1-5) + WAI-TECH-SF (12 items, 1-7) added to live Rina survey SV_734qjklvoJkl4to via additive API (no rebuild, QIDs/data preserved). Consent anonymity->confidential corrected. Published v21. API + participant-browser verified. FINAL Word + codebook regenerated.
2026-06-08: deadline -> 2026-06-10 (by אלעד)
2026-05-28: added after Inbal approved both instruments

Source

telegram-bot.db · created

Task #30 · CLOSED

Yuval prepares CLPsych 2026 invited talk slides (Article 8 Lia CPE)

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Yuval Holzman

Deadline 2026-05-31 (3 days)

Type / hard_deadline

Article #8 — CPE / CLPsych (Lia)

Last touched 2026-06-08

Description

Article 8 (Lia CPE methodology) was ACCEPTED at CLPsych 2026 + Yuval Holzman invited to give a talk (honour). Yuval needs talk slides ready by Sunday 2026-05-31 (week-of preparation window). Talk format / length TBD - confirm with CLPsych organizers.

History

2026-06-08: CLOSED — superseded by #33 (same CLPsych slides work; #33 has the locked-outline deliverable + Aya-driven 06-04 deadline)
2026-05-28: added after Elad reported CLPsych acceptance + invited-talk honour

Source

telegram-bot.db · created

Task #25

Dashboard health audit (weekly)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Claude/menti-agent (Telegram bot)

Type / recurring

Last touched 2026-07-02

Description

Recurring Sunday 9:00am Telegram health audit. Runs validate.py --report-mode and sends results to Elad. Defined in telegram-bot.db task id=25 with schedule "0 9 * * SUN".

History

2026-07-02: 2026-07-02: standing recurring Sunday audit (telegram bot, claim-before-run fix 06-07) - staleness expected for a recurring task
2026-05-27: created with dashboard sync architecture

Source

telegram-bot.db · created

Task #33 · CLOSED

CLPsych 2026: Yuval prepares slides + script from locked outline

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Yuval

Deadline 2026-06-04 (7 days)

Type / action_required

Article #8 — CPE / CLPsych (Lia)

Last touched 2026-06-08

Description

Outline locked 2026-05-31 PM with Elad. Yuval to convert outline into full slides + speaker script. Deadline driven by Aya's upload deadline (~2026-06-04).

History

2026-06-08: CLOSED — marked done from dashboard by אלעד
2026-06-08: 2026-06-08: OVERDUE - follow up. Owner confirmed: Yuval.
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #34 · CLOSED

CLPsych 2026: Claude records demo simulation video for talk

Status CLOSED 2026-05-31

סמן

עדכון תאריך

הערות

Owner Claude

Deadline 2026-06-03 (6 days)

Type / action_required

Article #8 — CPE / CLPsych (Lia)

Last touched 2026-05-31

Description

Demo Cesura sim built 2026-05-31. Claude to record an end-to-end run (audio + screen). Used in the invited talk + as visual reference for Hili's pilot. Deadline driven by Aya's upload deadline.

History

2026-05-31: CLOSED — Cesura URL provided by Elad 2026-05-31 PM: https://app.cesura.ai/sim/3978eb32-3e8b-407a-96b7-c7925a7a00ce — patched into LIVE PRE EOSRedirect via patch_cesura_url_2026-05-31.py. Internal docs updated. Site v15.14+ live with real URL.
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #35 · CLOSED

CLPsych 2026: Aya reply re exact video upload deadline (asked 2026-05-31, est 2026-06-04)

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Aya (CLPsych) / Elad waiting

Deadline 2026-06-02 (5 days)

Type / action_required

Article #8 — CPE / CLPsych (Lia)

Last touched 2026-06-08

Description

Email sent to Aya 2026-05-31. Reply expected to confirm the EXACT upload deadline. Currently estimated 2026-06-04 — but Aya's confirmation drives the Yuval slides + Claude recording sub-deadlines.

History

2026-06-08: CLOSED — marked done from dashboard by אלעד
2026-06-08: 2026-06-08: OVERDUE - follow up. Owner confirmed: Aya (CLPsych) / Elad waiting.
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #36

MENTI website: add contact form / details submission

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner TBD (web)

Type / action_required

Last touched 2026-07-02

Description

Reviewed mentiverse.ai 2026-05-31 — visitors cannot submit details (no contact form). Need to add one. Captures interest from researchers/practitioners who visit.

History

2026-07-02: 2026-07-02: parked - owner TBD; revisit when website work resumes
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #37 · CLOSED

Article 17 (Hili pilot): Yuval verifies Hili's experimental setup

Status CLOSED 2026-06-02

סמן

עדכון תאריך

הערות

Owner Yuval

Deadline 2026-06-02 (5 days)

Type / action_required

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-02

Description

Setup built 2026-05-31. Yuval to verify (probably at the 2026-06-02 13:00 dry-run Zoom or before). Final check before pilot 2026-06-07.

History

2026-06-02: CLOSED — Yuval reviewed Hili's setup at/around the 2026-06-02 dry-run. Final 3-researcher full-flow test rescheduled to after Hili's Thursday comments land (new task added).
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #38 · CLOSED

MENTI grant: budget breakdown verified vs Ministry of Science (uploaded to Dropbox)

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Elad

Type / action_required

Last touched 2026-06-08

Description

Reviewed 2026-05-31 — Elad entered the Ministry of Science portal and verified exact budget breakdown. Updated PDF saved to Dropbox. Closes out the budget verification step.

History

2026-06-08: CLOSED — marked done from dashboard by אלעד
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #39 · CLOSED

Evelyn (Article 3) budget: research actigraphy cost to finalize per-participant rate

Status CLOSED 2026-06-08

סמן

עדכון תאריך

הערות

Owner Elad

Type / action_required

Article #3 — Evelyn new run

Last touched 2026-06-08

Description

Discussed 2026-05-31. Estimate is 120-150 NIS per participant pending actigraphy cost. Elad to look into actigraphy quote (device rental or purchase) then finalize.

History

2026-06-08: CLOSED — marked done from dashboard by אלעד
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #40 · CLOSED

Article 17 (Hili pilot): Research protocol sent to Hili 2026-05-31

Status CLOSED 2026-05-31

סמן

עדכון תאריך

הערות

Owner

Type / action_required

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-05-31

Description

Hili now has the protocol. Confirms the alignment + dry-run prep on 2026-06-02.

History

2026-05-31: CLOSED — Closed at creation — captured as 'done' (we sent it; just logged here as audit trail).
2026-05-31: added via ops.py task add

Source

telegram-bot.db · created

Task #41 · CLOSED

Article 17: replace email with last-4-of-ID linker in Qualtrics PRE+POST + Cesura

Status CLOSED 2026-06-07

סמן

עדכון תאריך

הערות

Owner Elad + Yuval

Deadline 2026-06-06 (9 days)

Type / build

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-07

Description

Dry-run decision #1 (FIRM). Replace email field with a last-4-digits-of-national-ID linker as the manual PRE->sim->POST cross-survey key (no automation between Cesura and POST). Add the same 4-digit field to POST top + Cesura side. FLAG to Elad first: 4 digits not unique across 40 students - confirm whether to use last-4+birth-year / last-5-6 / keep ResponseId as primary. Update consent text (email->ID), codebook, For_Hili, start.html. menti-agent builds Qualtrics + propagates docs after Elad confirms field spec.

History

2026-06-07: CLOSED — Done 2026-06-05: last-4-ID linker added to PRE+POST (and Cesura asks last-4); confirmed working when pilot linked N=29 by 4-digit ID
2026-06-02: added via ops.py task add

Source

telegram-bot.db · created

Task #42 · CLOSED

Article 17: Yuval applies Hili's Thursday comments to SoT docx (track-changes)

Status CLOSED 2026-06-07

סמן

עדכון תאריך

הערות

Owner Yuval

Deadline 2026-06-05 (8 days)

Type / writing

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-07

Description

GATE B. After Hili sends phrasing comments Thu 2026-06-04 AM, Yuval applies them to Article_17_Questionnaire_02_0626 after Yuval.docx via Word track-changes, saves same filename, uploads to folder root. Covers: קלינית->טיפולית sweep; Scale C first-person rephrase; demographics edits (add children's-age, drop planning-to-be-parent, prior-clinical->prior-professional-engagement therapeutic/educational/counseling, add are-you-a-parent, Scale C item 6 relevant-to-everyone while preserving H4). Then menti-agent reads track-change XML + propagates to LIVE Qualtrics + codebook + built docs + site.

History

2026-06-07: CLOSED — Done 2026-06-04/05: Hili's comments applied to SoT docx; clean final questionnaire built; propagated to LIVE PRE v29/POST v25
2026-06-02: added via ops.py task add

Source

telegram-bot.db · created

Task #43 · CLOSED

Article 17: DESIGN curiosity measure (סקרנות) for Scale C pre+post

Status CLOSED 2026-06-07

סמן

עדכון תאריך

הערות

Owner Elad + Yuval

Deadline 2026-06-05 (8 days)

Type / design

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-07

Description

Dry-run decision #4 - NEEDS ELAD DESIGN DECISION (not just Hili's wording). Hili's central goal is increasing curiosity toward parent-work; Elad agreed to add it pre+post. NEW construct. Elad decides: (1) custom Yoav-style 1-10 items vs validated curiosity scale; (2) item count (1-3 keeps 45-min budget); (3) hypothesis status - primary/secondary/exploratory, possible H8 with Holm correction; (4) inside Scale C matrix vs separate block. Then Yuval -> SoT docx -> menti-agent propagates to Qualtrics + codebook + For_Hili hypotheses + protocol + R plan.

History

2026-06-07: CLOSED — Done: curiosity (סקרנות) perceived-change item added pre+post (1-5); live in pilot data (perceived_change curiosity M=3.83, 66% increased, p<.001)
2026-06-02: added via ops.py task add

Source

telegram-bot.db · created

Task #44 · CLOSED

Article 17: Elad opens shared WhatsApp group with Hili

Status CLOSED 2026-06-02

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-04 (7 days)

Type / coordination

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-02

Description

Dry-run decision #8c. Hili asked for a small shared WhatsApp group (Hili+Yuval+Elad) for ongoing coordination; gave her number at the meeting. Elad opens it. menti-agent does NOT open/send. Also (#8d) Elad checks the college-sent email format + makes format fixes (Elad already started inquiries).

History

2026-06-02: CLOSED
2026-06-02: added via ops.py task add

Source

telegram-bot.db · created

Task #46 · CLOSED

Article 17: 3-researcher full-flow end-to-end test (PRE->Cesura->POST)

Status CLOSED 2026-06-07

סמן

עדכון תאריך

הערות

Owner Elad + Yuval + Hili

Deadline 2026-06-06 (9 days)

Type / testing

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-06-07

Description

Dry-run decision #8b. All three run the complete flow as test subjects to confirm 'the water flows through the pipes' end-to-end. Meeting moved this to AFTER Hili's Thursday comments land (so the test exercises the FINAL version). BLOCKED ON: last-4-of-ID change in Qualtrics+Cesura + Hili's Thursday comments + Yuval SoT docx update propagated to LIVE. Hard constraint per Elad: stable end-to-end before pilot 2026-06-07 12:00.

History

2026-06-07: CLOSED — Done: the pilot itself was the live end-to-end run (PRE->Cesura Lia->POST); flow worked, N=29 linked
2026-06-02: added via ops.py task add

Source

telegram-bot.db · created

Task #47

Article 17 Hili pilot - Elad reviews preliminary report + decide on powered/controlled follow-up (Year-2 plan)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type / analysis

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-07-02

Description

Pilot done + analyzed 2026-06-07 (N=29, both primaries supported). Password-protected report shared (https://elad-refoua.github.io/article17-hili-pilot/results/, pw LiaPilot2026). Decide: write up as feasibility/acceptability pilot now, and/or scale to a powered controlled follow-up (Year-2: 80 students, quasi-natural comparison).

History

2026-06-07: added via ops.py task add

Source

telegram-bot.db · created

Task #48 · CLOSED

Article 5 Yoav -> Cesura: build sim from 18-field doc + wire PRE->sim->POST

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Yuval

Deadline 2026-06-15 (18 days)

Type / build

Article #5 — Socratic Yoav

Last touched 2026-07-02

Description

IN PROGRESS (2026-06-15): Yuval is uploading the Yoav simulation to Cesura. HIGH (track: therapists). Build the Yoav Cesura sim from the ready 18-field doc (study/articles/5.yoav/Prompt/בוט יואב - מבנה שדות סזורה Cesura v2 2026-06-03.docx): paste all 18 fields, set Guide=Off, then wire PRE->sim->POST redirect chain.

History

2026-07-02: CLOSED — DONE: Yoav live on Cesura - Friday 2026-06-26 run: 20 started / 8 completed / 12 partial; Danny Derby first run 06-27
2026-06-09: Meeting 2026-06-08 (missed by auto-summary, caught in transcript review): after the Cesura upload, hand the Yoav sim to Danny Derby to run another cohort; sequenced after staffing (Moria/RA onboarding).
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #49

Article 3 Evelyn - build 6 multi-session prompts + deploy to Cesura + engagement

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / build

Article #3 — Evelyn new run

Last touched 2026-07-02

Description

track: parent. Yuval builds the 6 multi-session Evelyn prompts + deploys to Cesura + engagement setup. IRB already cleared (Israel-only, no UCL). Budget = task #39.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #50

Article 14 Clalit - Clalit sent an email (2026-06-15): update the Clalit proposal accordingly (owner Elad)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad + Yuval

Deadline 2026-06-30 (33 days)

Type / action_required

Article #14 — Clalit / Dalia

Last touched 2026-07-02

Description

track: parent. TRIGGER: Tziki Shibak (Tzvika, zvish4@clalit.org.il, Research Promotion / Clalit Center District) sent 2 emails 2026-06-14 - the final small round before Clalit DISTRICT approval. Requested protocol fixes: (a) the marking field '[yuktsa al yedei Clalit]' -> write the Helsinki number when received OR remove entirely; (b) list the Clalit contact person = PI Dalia Kelly Benacot (Dalia Kelly Benacot). Then return the corrected protocol TOGETHER WITH the attached forms (tofes bakasha lebitsua mehkar; ktav hitchayvut sodiyut; nohal hotsaat netunim mehavada hamehozit; nispach lehagasha 1 - no need to fill research details there) -> district approval -> Helsinki committee. Builds on 'protocol mehkar Menti - Clalit - metukan 2026-06-08.docx'. Owner: Elad does the corrections himself. Admin/aggregate only - no patient data.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #51

MENTI Study Studio - Yuval review + schedule walkthrough with Elad

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / review

Last touched 2026-07-02

Description

track: methodology. Yuval reviews the new MENTI Study Studio (חבילת השת״פ) tool + schedule a walkthrough with Elad. Live: https://elad-refoua.github.io/study-studio/ ; folder study/MENTI Study Studio/.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #52

Recruit Research Assistant (RA) for MENTI - flyer published 2026-06-26; track applications/responses

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Deadline 2026-07-15 (48 days)

Type / action_required

Last touched 2026-07-02

Description

HIGH (track: methodology). Recruit an RA for the MENTI project: post the recruitment ad, screen candidates, onboard. Ad drafted by the writer agent this session.

History

2026-06-08: deadline -> (cleared) (by selftest)
2026-06-08: deadline -> 2026-07-15 (by Elad)
2026-06-08: ????? ????? ????????? (note by selftest)
2026-06-08: REOPENED — reopened from dashboard by selftest
2026-06-08: CLOSED — marked done from dashboard by selftest
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #53

Article 9 (CPE vs flat-prompt LIWC) - deploy CPE Alice on Cesura + recruit ~16 therapists (protocol + prompts + IRB all ready)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad + Yuval

Type / build

Article #9 — CPE vs Flat Prompt LIWC

Last touched 2026-07-02

Description

Protocol + both CPE prompt sets (Alice x3, Luke x3) written; IRB cleared via Lavidor 2026-05-15 under MENTI 290125444. Remaining: deploy CPE Alice (GAD) on Cesura, pilot 2-3 therapists, then recruit ~15-16 mental-health professionals for single English sessions; transcripts auto-saved for LIWC-22 comparison vs published Client101 data. No Cesura deployment and no data yet.

History

2026-06-08: Meeting 2026-06-08: design = CPE Alice + Luke vs the original published JMIR prompts (show CPE yields deeper/richer output); prepare a pre-registration.
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #55

Article 13 (Prompt Compression) - launch rater recruitment + open blinded expert evaluation (platform live, stimuli + IRB ready)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad + Yuval

Type / build

Article #13 — Prompt Compression

Last touched 2026-07-02

Description

Prompt Compression - blinded expert evaluation. RATER PORTAL: https://elad-refoua.github.io/prompt-compression-eval/ | ADMIN (issue per-rater links + watch responses): https://elad-refoua.github.io/prompt-compression-eval/admin.html (repo elad-refoua/prompt-compression-eval). 20 Hebrew stimuli (10 matched 33K/3K pairs) + recruitment package (yuval-package/) ready; IRB cleared via Lavidor 2026-05-15 (Amendment 1, MENTI 290125444). PRE-LAUNCH CHECK: confirm the backend API URL is wired in config.json - the portal renders but rating submissions fail SILENTLY without it. Then: issue per-rater URLs from admin.html, send recruitment, collect blinded ratings from 3-5 expert clinical psychologists (no rater responses yet).

History

2026-06-08: Meeting 2026-06-08: both-bots arm CONFIRMED (coordinator's suggestion); also review prompts/protocol toward a journal pre-registration.
2026-06-08: PENDING from Yuval meeting 2026-06-08 (to confirm after transcript): Yuval suggests adding a condition where therapists talk with BOTH bots and say/judge which is which.
2026-06-08: Added rater portal + admin links + pre-launch API-URL-wiring check to the card (Elad request, 2026-06-08).
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #56

Article 13: add research arm - therapists converse with BOTH Sara bots (33K vs 3K) themselves + judge, beyond rating the recorded portal sessions

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / build

Article #13 — Prompt Compression

Last touched 2026-07-02

Description

Meeting 2026-06-08: add an arm where therapists interact directly with both bots (long 33K vs short 3K prompt) and compare, in addition to the existing blinded rating of recorded conversations in the live portal. Fold into the pre-registration.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #57

Evelyn study: build detailed validated-questionnaire table (items, languages, codes) from the Questionnaires folder

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / build

Article #3 — Evelyn new run

Last touched 2026-07-02

Description

Meeting 2026-06-08: blocks the Qualtrics build for Evelyn. Compile from the existing Questionnaires folder.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #58

Evelyn study: consult Karen on shortening/merging questionnaires to cut participant burden

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad + Yuval

Type / action_required

Article #3 — Evelyn new run

Last touched 2026-07-02

Description

Meeting 2026-06-08 streamlining: drop pre-test in session 1, merge FQ9 into session-3 post, add a 1-month follow-up. Confirm with Karen.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #59

Fix Dropbox sync - coordinator files arrive EMPTY for Elad (technical)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / action_required

Last touched 2026-07-02

Description

Meeting 2026-06-08: wasted meeting time hunting empty files; check shared-folder sync / Dropbox support.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #60

Confirm Moria (overseas RA) arrival dates + plan intensive MENTI integration

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / action_required

Last touched 2026-07-02

Description

Meeting 2026-06-08: schedule Moria into MENTI work once arrival is known.

History

2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #61 · CLOSED

Sunday meeting with Nadav re budgets + university finance mechanisms; duplicate lab key for the coordinator

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad

Deadline 2026-06-14 (17 days)

Type / action_required

Last touched 2026-07-02

Description

Meeting 2026-06-08: finance/budget walkthrough with Nadav; copy a lab key.

History

2026-07-02: CLOSED — marked done from dashboard by אלעד
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #62 · CLOSED

Article 10 (Narrative Review): Wed 2026-06-10 work meeting with Zohar to finalize + submit the review

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad + Yuval + Zohar

Deadline 2026-06-10 (13 days)

Type / hard_deadline

Article #10 — Narrative Review (5-Axis)

Last touched 2026-07-02

Description

Meeting 2026-06-08: final closure of the narrative-review paper (Article 10, Yuval-led). Wed work meeting with Zohar; submit target 2026-06-10. (Earlier mislabeled 'Nature article' by the auto-summary.)

History

2026-07-02: CLOSED — Superseded by task 67 - Article 10 closing tracked there (v30 with Yuval since 06-21)
2026-06-09: Corrected: this is Article 10 (Narrative Review), NOT Nature; repointed article_ref->10; owner Elad+Yuval+Zohar.
2026-06-08: added via ops.py task add

Source

telegram-bot.db · created

Task #63

Article 17 (Hili): review v3 manuscript + decide Scale B status / length / author lists

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type / review

Article #17 — Hili Pilot — MentiTherapist

Last touched 2026-07-02

Description

v3 manuscript is AUDIT-PASSED (2 deep reviews + Crossref) + Word sent. Elad does line-by-line review (USER-LOCK). 3 open decisions surfaced by the deep review: (1) Scale B apprehension co-primary vs low reliability (alpha .39/.62) - keep / demote / swap validated measure; (2) length ~6300 words vs Brief-Report 4-5k - trim or confirm outlet tolerates; (3) author lists (main + Article 5 Refoua et al. in prep) - fill placeholders before submission. Target: educational/school-psych Q1 Brief Report (School Psychology Review; alt Training and Education in Professional Psychology).

History

2026-06-09: added via ops.py task add

Source

telegram-bot.db · created

Task #64 · CLOSED

Article 15 (Rina): decide participant identifier - 4-digit T.Z. cross-link (Cesura + questionnaire) vs keep email; consult Inbal on field feasibility

Status CLOSED 2026-06-14

סמן

עדכון תאריך

הערות

Owner Elad + Inbal

Type / decision

Article #15 — Rina

Last touched 2026-06-14

History

2026-06-14: CLOSED — RESOLVED 2026-06-14: identifier decided with Inbal -- keep mandatory email (QID3) + 'חסוי'/confidential consent; v21/v23 retain email. No 4-digit T.Z. cross-link.
2026-06-14: Open decision (Elad reconsidered identifier 2026-06-14, consults Inbal himself). Live v21 currently KEEPS the mandatory email + consent now reads 'chasui'/confidential. Lead idea: 4-digit T.Z. entered in BOTH Cesura Rina sim AND questionnaire, cross-linked (same last-4-ID pattern as Hili/Article 17). Inbal runs sim in field - her input on whether students reliably enter same 4 digits in both places is the gate. No live-survey change made.
2026-06-14: added via ops.py task add

Source

telegram-bot.db · created

Task #65 · CLOSED

Article 15 Rina - prepare Cesura Learning Space for the Rina simulation (final run Thu 2026-06-18, paired with questionnaire)

Status CLOSED 2026-07-02

סמן

עדכון תאריך

הערות

Owner Elad + Yuval/Inbal

Deadline 2026-06-18 (21 days)

Type / build

Article #15 — Rina

Last touched 2026-07-02

Description

Prepare a Cesura Learning Space (מרחב עבודה) for the Rina simulation so it can be deployed to participants. The final/last sim run is Thursday 2026-06-18, run together with the v23 questionnaire (this is effectively the task #6 pilot 5-10 run). Questionnaire send tomorrow (2026-06-15) is gated on Yuval's review pass. Pair the Learning Space deployment with the questionnaire link.

History

2026-07-02: CLOSED — marked done from dashboard by אלעד
2026-06-15: added via ops.py task add

Source

telegram-bot.db · created

Task #67 · CLOSED

Finish Article 10 (narrative review) then Elad+Yuval meet

Status CLOSED 2026-07-06

סמן

עדכון תאריך

הערות

Owner Yuval Holzman

Deadline 2026-06-30 (33 days)

Type / writing

Article #10 — Narrative Review (5-Axis)

Last touched 2026-07-06

Description

Yuval finishes the Article 10 (5-axis narrative review) by Tue 2026-06-30; then Elad + Yuval meet to close it. Work from v30 (MANUSCRIPT_AdvSimulation_v30_BodyCitations_2026-06-21.docx).

History

2026-07-06: CLOSED — Article 10 finished + SUBMITTED to JMIR Medical Education 2026-07-06 (submission #106300). Manuscript complete; now under review. Elad+Yuval close-out meeting no longer gated on 'finishing' — article is out the door.
2026-06-26: added via ops.py task add

Source

telegram-bot.db · created

Task #68

Organize Evelyn (Article 3) questionnaires in BOTH Hebrew and English

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval Holzman

Type / build

Article #3 — Evelyn new run

Last touched 2026-07-02

Description

Yuval organizes all of Evelyn's (Article 3) questionnaires in both Hebrew and English versions.

History

2026-06-26: added via ops.py task add

Source

telegram-bot.db · created

Task #69

Learn Prolific + the Evelyn-on-Prolific run protocol

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval Holzman

Type / research

Article #3 — Evelyn new run

Last touched 2026-07-02

Description

Yuval gets to know the Prolific platform and the Evelyn (Article 3) on-Prolific run protocol we already built.

History

2026-06-26: added via ops.py task add

Source

telegram-bot.db · created

Task #70

Yuval learns (with her own Claude) to connect Claude Code to GitHub

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / build

Last touched 2026-07-02

Description

Build (in Claude Code) a capability to talk to GitHub, intended for Yuval to use.

History

2026-06-26: 2026-06-26: reframed per Elad — Yuval learns this herself using her Claude (not Claude-builds-for-her).
2026-06-26: added via ops.py task add

Source

telegram-bot.db · created

Task #71

Yuval learns (with her own Claude) to build a questionnaire-building skill, based on Elad's

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Yuval

Type / build

Last touched 2026-07-02

Description

Build a questionnaire-building skill for Yuval, based on Elad's existing questionnaire-building skill/workflow.

History

2026-06-26: 2026-06-26: reframed per Elad — Yuval learns this herself using her Claude.
2026-06-26: added via ops.py task add

Source

telegram-bot.db · created

Task #74

UP big paper (12B) -- submit full study to JMIR Medical Education (npj deadline dropped)

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type / action_required

Article #122 — UP Within-Subjects (Paper #2)

Last touched 2026-07-15

Description

Full study to JMIR Medical Education (Elad first author; Nature Mental Health option open). Study built from ZOHAR's IPT protocol (Zohar sent it). Research plan written (docs/STUDY2_FULL_RESEARCH_PLAN_2026-07-15), per-module-first analysis order. STATUS (2026-07-15, Elad): (1) v6.1 cleaning (N=76/711) -- Shaked has FINISHED his code review; FINAL JOINT Elad+Shaked review -> USER-LOCK scheduled for Thu 2026-07-16. (2) PRE-REGISTRATION -- SCOPE RESOLVED (Elad 2026-07-15): fresh pre-reg to be PREPARED FROM SCRATCH FOR 12B -- this is the active upstream pre-reg deliverable to build (written from STUDY2_FULL_RESEARCH_PLAN_2026-07-15). NO finalized/approved/registered pre-reg for 12B; prior v4.0/v3.4 = superseded early drafts, never registered. Fresh pre-reg IN PREPARATION; OSF click pending Elad. (Article 18 has its own separate working draft -- see Task #75.) (3) Barlow co-authorship approached at the END, after a draft exists. THEN: finalize fresh pre-reg + analysis code (r-coder) -> run -> manuscript (writer).

History

2026-07-15: 2026-07-15: correction (Elad) -- no finalized/approved/registered pre-reg for 12B; being redone from scratch; v4.0/v3.4 = superseded early drafts. Redo scope = confirm with Elad.
2026-07-15: 2026-07-15: reworded per Elad -- v6.1 USER-LOCK in joint Elad+Shaked code review; decisions+pre-reg in prep by Elad/team; Barlow deferred-to-post-draft confirmed.
2026-07-15: 2026-07-15 (menti-agent): added concrete upstream steps; recorded Study-2 plan written + analysis restructured. No deadline (npj dropped).
2026-07-06: added via ops.py task add

Source

telegram-bot.db · created

Task #75

UP lecture study (Article 18) -- build: questionnaire + Cesura workspace + flyer + Zohar consult

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Shaked (questionnaire+workspace+recruitment), Omer (deliver lecture), Elad (flyer+Zohar)

Type / build

Article #18 — UP Lecture Study

Last touched 2026-07-15

Description

Build the standalone lecture study (Article 18). PRE-REGISTRATION (SCOPE RESOLVED, Elad 2026-07-15): Art.18 ALREADY HAS a working draft -- a Study-1 protocol draft + an EARLY WORKING AsPredicted pre-reg DRAFT (docs/STUDY1_M3_LECTURE_PROTOCOL_2026-07-14, docs/STUDY1_M3_LECTURE_PREREG_ASPREDICTED_2026-07-14) with open decisions still listed. NEITHER finalized, the pre-reg NOT registered on AsPredicted -- but the draft EXISTS. Art.18 is NOT the record being prepared-from-scratch (that is 12B). Do NOT treat this draft as approved/registered; AsPredicted-registration click stays pending Elad. Owners: Shaked = dedicated questionnaire + NEW Cesura workspace + Facebook recruitment (cap ~100); Omer = prepare + deliver the ~90-min M3 lecture; Elad = flyer + consult Zohar (IPT-evenings design + reviewer lessons). Target ~34+ participants for a medium effect. Target journal Cognitive Therapy & Research (or similar).

History

2026-07-15: 2026-07-15: correction (Elad) -- the 07-14 pre-reg is an early working draft, not finalized/registered. Fresh-redo scope for Art.18 = confirm with Elad.
2026-07-15: 2026-07-15 (menti-agent): added concrete detail + owners; recorded Study-1 protocol + pre-reg drafts written 2026-07-14.
2026-07-06: added via ops.py task add

Source

telegram-bot.db · created

Task #76

Zvia Zeligman bot study — planned separate bot study (Elad + Zvia); assign article #

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type / planning

Last touched 2026-07-15

Description

Elad update 2026-07-15: a separate bot study Elad + Zvia Zeligman wanted to build ('mechkar acher im Ziviya Zeligman shel bot she-ratzinu la-asot'). GATED ON YUVAL HOLZMAN's RETURN (to be reset/re-baselined when Yuval is back). RECONCILE for Elad: this is almost certainly the existing 'Zivia' collaboration already on record = the Window of Tolerance / CPTSD dual-agent psychoeducation bot + grant proposal (meeting study/Meetings/zivia - cptsd menti 251225/, Dec 2025) — confirm same person (spelling Zivia vs Zvia Zeligman) and same study. FLAG: needs a formal article # assigned (do NOT reuse the A16 Zimbardo stub). Person added to people map (collaborator, bot study).

History

2026-07-15: added via ops.py task add

Source

telegram-bot.db · created

Task #77

Article 15 Rina — schedule a meeting with Inbal Cohen about Rina's bot

Status OPEN

סמן

סדר עדיפות

עדכון תאריך

הערות

Owner Elad

Type / coordination

Article #15 — Rina

Last touched 2026-07-15

Description

Elad update 2026-07-15: a meeting with Inbal Cohen (David Yellin field coordinator) needs to be scheduled about Rina's bot. Elad to set it up. Context: control-arm recruitment (SV_3CQw2i9BhfgMuAm) is the bottleneck (4 real responses).

History

2026-07-15: added via ops.py task add

Source

telegram-bot.db · created

Article #1 · MentiParent (Danny + Evelyn)

The MentiParent chatbot: An AI-based approach to enhancing parental reflective functioning

Status ACCEPTED Parent

Status note ACCEPTED — Karen final proofing (2026-04-17)

Authors Yirmiya, K.*, Refoua, E.*, Truscott, A., Reeve, H., Fonagy, P., Elyoseph, Z.

Journal Scientific Reports (IF 3.9)

Data N=60 practitioners (Danny) + N=36 parents (Evelyn). Feasibility M=5.61, Acceptability M=5.56.

Folder study/articles/1.danny/ open

Last touched 2026-07-02

Verify ⚠ Verification needed

Next step

Verify Scientific Reports production/publication status + collect proofs (still pending as of 2026-07-02)

Notes

Per audit row 33: VERIFY NEEDED — 40 days since 'Karen final proofing' note. Most likely already published or close.

History

2026-05-28: status confirmed (accepted, Karen final proofing)
2026-04-17: final proofing

Article #2 · Foster Care (Julie/Guli)

Fostering Reflection: Development and Initial Evaluation of a Mentalization-Based GenAI Simulator for Foster Care Supervisors

Status UNDER REVIEW Foster

Status note UNDER REVIEW — no response yet from journal (2026-04-17)

Authors Elyoseph, T.*, Refoua, E., Novis-Deutsch, N., Yirmiya, K., Fonagy, P., Enosh, G.

Journal Child Abuse & Neglect

Data N=46 female foster care supervisors. Recommend M=6.67 (93.4% scored 6-7). Feedback M=5.83.

Folder study/articles/2.Julie/ open

Last touched 2026-07-02

Verify ⚠ Verification needed

Next step

Ping Tal - reviewer silence since May; request journal status (refreshed 2026-07-02)

Notes

Per audit row 34: VERIFY NEEDED — worth a status ping with Tal.

History

2026-05-28: status confirmed (under review at Child Abuse & Neglect)

Article #3 · Evelyn new run

Evelyn Multi-Session Training Study

Status BUILDING Parent

Status note IRB FULLY CLEARED (Israel-only, no UCL needed) — Yuval BUILDING

Authors Refoua / Yirmiya / Yuval

Data Design LOCKED 2026-04-29. 3 weekly sessions x Evelyn-girl + Evelyn-boy = 6 sims. Yuval building 6 prompts.

Folder study/articles/3.Evelyn/ open

Last touched 2026-07-15

Verify ⚠ Verification needed

Next step

Design locked (mtg 2026-06-08): 3 bots/sessions; ADD SDQ + Parenting Self-Efficacy pre+post; 1-month follow-up. NEW directions (2026-07-01/02): run via PROLIFIC for efficient recruitment; possible RCT with a no-simulation CONTROL arm; open deliberations on study length + usage; consult KAREN on shortening/merging the questionnaires. Blocked on detailed questionnaire table (#57) + Karen consult (#58); Yuval organizes all questionnaires in Hebrew + English (#68).

Notes

BIU IRB via Prof. Michal Lavidor covers Articles 3, 7, 9, 13 (single approval). Per audit row 35: VERIFY NEEDED — Yuval prompt-build status.

History

2026-07-15: 2026-07-15: Elad update — Evelyn (A3) is GATED ON YUVAL HOLZMAN's RETURN (coordinator). To be reset/re-baselined when Yuval is back.
2026-07-02: 2026-07-02: new directions - Prolific recruitment + possible RCT control arm + length/usage deliberations + Karen questionnaire consult (briefed to Karen via WhatsApp).
2026-06-26: 2026-06-26: Yuval to organize Evelyn questionnaires HE+EN (#68) + learn Prolific run protocol (#69).
2026-06-09: Clarification: 'Geffen Collins' in the 2026-06-08 entries = Yuval Holzman (Timeless voice mislabel).
2026-06-08: Meeting 2026-06-08: Evelyn design decisions (SDQ + PSE pre/post, streamlining, 1-mo follow-up). Geffen Dropbox sync issue flagged.
2026-05-15: IRB cleared (Lavidor)

Related tasks

#39 #49 #57 #58 #68 #69

Article #4 · Reimagining Clinical Training (Carol+Dweck triad)

Reimagining Clinical Training in the AI Era: Generative Simulation as Scaffolding for Therapist Skills

Status IN PREP Therapists

Status note IN PREP — Carol/Jeff/Dafna framework triad

Authors Refoua, E. (1st), Bar, M., Linkovski, O., Elyoseph, T., Elyoseph, Z., Yirmiya, K., Rafaeli, A.K., Rafaeli, E.

Data NMS Training Agents on Elad's OWN platform (NOT Cesura): Carol (needs) LIVE [carol bot -updated 260626.docx; + 'End conversation' (סיים שיחה) button instruction + existing session_termination hard-stop]; Dafna (stances) LIVE [Dafna_Bot_V8_2026-06-25.docx; V7=Carol-style conversational overhaul + gradual opening + Gloria/Perls clip; V8=session_termination HARD-STOP + 'End conversation' button instruction]; Modes='Maya' first-draft pending Eshkol sign-off. Reusable socratic-training-bot skill (TDD-verified) available to build the rest.

Folder study/articles/4. Dr Carol and guidance/ open

Last touched 2026-07-15

Next step

Carol + Dafna LIVE on Elad's platform (Carol 260626; Dafna V8). Modes/Maya draft pending Eshkol sign-off. socratic-training-bot skill available to build the rest. Carol still awaiting Eshkol & Yuval feedback (sent 2026-06-21).

Notes

AUDIT FIX (row 36): Removed stale 'Sunday 2026-05-03 — Yuval represents Elad' next-step phrasing (24 days stale). Replaced with VERIFY NEEDED. Task #4 status_check tracks this.

History

2026-07-15: 2026-07-15: Elad update — 'Dr. Carol's study' + 'Dr. Daphna's study' are GATED ON YUVAL HOLZMAN's RETURN (coordinator); to be reset/re-baselined when Yuval is back. Elad corrected the name to 'Dr. Carol' (NOT Kron/Karen). FLAG for Elad: in current records Carol (Dr. Dweck=NEEDS) and Dafna (POSITIONS) are the Article-4 training-BOTS coordinated by Yuval+Dafna — confirm whether 'Dr. Carol's study' and 'Dr. Daphna's study' ARE Article 4, or separate person-led studies needing their own article #s.
2026-06-26: 2026-06-26: Dafna finalized V6->V7->V8 + Carol updated to 260626; Carol + Dafna LIVE on Elad's own platform (not Cesura).
2026-06-25: 2026-06-25: socratic-training-bot skill built (TDD RED->GREEN, 11/11 PASS, gold=Carol); Dafna stance-bot cleaned V3->V6 deployable (refs stripped, exact stage arc, always-present side-window clip) - Elad uploading to his own platform; Maya/modes first-draft pending Eshkol.
2026-04-17: Carol bot updated
2026-05-03: Yuval-Eshkol meeting (outcome unrecorded)

Related tasks

Article #5 · Socratic Yoav

Socratic Dialogue Training (Yoav CPT) — two-study paper

Status RECRUITING Therapists

Status note RECRUITING - OSF pre-reg fz8xm LIVE since 2026-05-28; new CPT cohort active; weekly response monitoring

Authors Refoua, E.* (1st), Linkovski, O., Rafaeli, E., Elyoseph, Z., Yirmiya, K., Derbi, D.

Data Live survey SV_039jMzxfdPX6yRU (share link ALWAYS as ...?new=1). FIRST Cesura run happened Fri 2026-06-26 (Danny Derby group): 20 started / 8 COMPLETED (finished=100%) / 12 not finished (6 at ~4%, 6 engaged at 26-35% = did PRE+sim, missed POST). 354->362 auditable. All 8 completers carry new='1'.

Folder study/articles/5.yoav/ open

Last touched 2026-07-02

Attention ★ Action needed this week

Next step

Study 2 confirmatory sample = new=1 responses 2026-06-26 onward (exclude CalStudio + QA test qa.verify@example.com). Email IS collected (QID45, ForceResponse=ON) so the 6 engaged drop-offs (26-35%) are remindable; Save&Continue=ON keeps partials 2 weeks -> reopening the SAME ?new=1 link on the SAME device resumes to the POST. OPEN/awaiting Elad's go (NOT done): reminder to the 6 to finish the POST. Then pull a dated new=1 breakdown + run the pre-registered Challenge-subscale (items 4/6/9) paired t per OSF fz8xm.

Notes

AUDIT FIX (row 37): Chip changed from 'Decision Point' to 'PAUSED'. Feasibility-only recommendation removed (paper is two-study, not feasibility-only). v18 Method+Results draft READY at Paper_Package_2026-05-18/. Study 2 pre-reg OSF fz8xm awaiting public approval ~2026-05-27. Iron rule: all numbers from code (v18_compute_all.R + v18_psychometrics.R -> v18_stats.json), not memory.

History

2026-07-02: 2026-07-02: Friday 2026-06-26 first Cesura run -> 20 started / 8 completed / 12 partial; all completers new=1; email collected (QID45) so 6 engaged drop-offs remindable; ?new=1 standing rule.
2026-06-26: 2026-06-27: Danny Derby runs Yoav on Cesura for the FIRST time = first real Cesura run / Study 2 confirmatory cohort begins (via Danny Derby's CPT channel).
2026-06-25: 2026-06-25: BOT-link href fixed to Cesura by Elad + verified as participant on the live form (was mis-wired: text=Cesura, href=CalStudio).
2026-06-25: 2026-06-25: Yoav platform switch CalStudio->Cesura recorded; Study 2 run starts 2026-06-26; flagged BOT-link href still CalStudio.
2026-06-15: 2026-06-15: Yuval uploading the Yoav sim to Cesura
2026-05-28: Yoav back to running - OSF live, Task #13 closed, monitoring task #12 reframed to active
2026-05-28: returned to active recruitment after OSF fz8xm went live
2026-05-18: N=49 re-analysis
2026-05-18 PM: subscale FA (Challenge d=0.38)
2026-05-20: v18 discovery-narrative
2026-05-21: Study 2 pre-reg ready
2026-05-25: OSF fz8xm SUBMITTED + paper PAUSED

Related tasks

#3 #11 #12 #13 #48

Article #6 · AI Rater

AI Rater for Socratic Fidelity (SDRS)

Status IN PREP Methodology

Status note IN PREP — Prompts v1.1, expert review pending

Authors TBD

Data 28 pipeline-ready transcripts from 48 conversations, 38 trainees. 5-pass multi-prompt system.

Folder study/articles/6. check socratic fidelity/ open

Last touched 2026-07-02

Next step

Expert review of prompts -> AI rating run -> calibration (unchanged; reviewed 2026-07-02)

Notes

Per audit row 38: VERIFY NEEDED — card hasn't moved in months.

Article #7 · Or experts

Or Foster Child Bot — Expert Evaluation

Status IN PREP FosterMethodology

Status note IN PREP — IRB cleared (Lavidor 2026-05-15); Tal analyzing

Authors Tal Elyoseph (lead)

Data 59 raw Qualtrics responses (SV_6kZsW4Ky4nPndno). Tal started analyzing. Expert platform live.

Folder study/articles/7. Or for experts/ open

Last touched 2026-07-02

Verify ⚠ Verification needed

Next step

Get fresh status from Tal - expert eval stalled since April; chase (refreshed 2026-07-02)

Notes

AUDIT FIX (row 39): Removed stale 'WhatsApp to Tal 2026-04-24 09:10' next-step. Replaced with VERIFY NEEDED. IRB cleared via Lavidor 2026-05-15.

History

2026-05-28: 2026-05-28 Elad: article spans foster + methodology tracks
2026-05-15: IRB cleared (Lavidor)

Article #8 · CPE / CLPsych (Lia)

Clinical Prompt Engineering: Encoding Clinical Knowledge into AI Simulations for Parent Training

Status ACCEPTED Methodology

Status note CLPsych 2026 ACCEPTED -- Yuval Holzman invited for a talk

Authors Holzman, Y. (1st), Refoua, E., Elyoseph, Z., Yirmiya, K., Elyoseph, T., Linkovski, O., Rafaeli, E.

Journal CLPsych 2026

Data Lia Track 1: 132 sessions, 42 completed. Safety M=5.80, Realism M=4.60.

Folder study/articles/8. Lia for CPE/ open

Last touched 2026-07-05

Next step

Yuval prepares CLPsych 2026 slides + script from the locked outline (2026-05-31); reviewed 2026-07-02

Notes

Status reverted 2026-05-15 from 'review passed' (inferred) to 'Under Review' (Elad correction). Yuval Holzman is research coordinator leading this paper. | 2026-05-28: ACCEPTED at CLPsych 2026 -- Yuval Holzman invited for an invited talk (significant honour).

History

2026-07-05: 2026-07-05: Yuval PRESENTED at CLPsych 2026 (oral) - excellent delivery, good audience reactions, and she won an excellence scholarship for the conference presentation. Companion page: sim link removed post-conference (page now safe for social linking); Facebook post drafted (writer agent) explaining CPE + Lia + the war-origin story, links in first comment
2026-07-02: 2026-07-02: CLPsych companion page FINAL (fullscreen zoom incl. 3x; Try-Lia open ALL conference day); English Lia sim link (8c5a5780) wired in + LIVE + smoke-tested as participant; team test request sent ONCE to menti-rachav group before Sat night
2026-05-31: 2026-05-31 PM: CLPsych talk prep session — outline locked + demo sim built (recorded version for talk + Hili copy) + email to Aya re upload deadline 2026-06-04
2026-05-28: CLPsych talk slides task #31 added (due Sun 2026-05-31)
2026-03-30: v16 manuscript file
2026-05-09: Elad uploaded revisions
2026-05-28: ACCEPTED + Yuval invited for invited talk

Related tasks

#30 #33 #34 #35

Article #9 · CPE vs Flat Prompt LIWC

Does Prompt Architecture Matter? Comparing CPE to Standard Prompting via Psycholinguistic Analysis

Status READY TO DEPLOY Methodology

Status note PROTOCOL READY — IRB cleared; ready to deploy Alice & Luke

Authors TBD

Data Protocol written; CPE prompts for Alice (GAD) and Luke (Depression) built (25,114 words total).

Folder study/articles/9. CPE vs flat prompt LIWC/ open

Last touched 2026-07-02

Next step

English-session constraint noted (briefed to Karen). Deployment not yet done (confirmed 2026-06-21); IRB cleared, awaiting deployment. WATCH: 9<->13 merge option under consideration (see notes) - Elad leaning keep separate for now.

Notes

IRB via Lavidor covers Articles 3/7/9/13. OPEN STRATEGIC OPTION (Elad 2026-07-01, cross-linked to #13, NOT a decision): merge #9+#13 into one 3-arm within-subjects study - (1) flat/Client101 baseline, (2) full CPE ~33K, (3) compressed CPE ~3K; 10 therapists each run all 3 (+optionally AI agents); measured by BOTH LIWC (objective) AND clinician richness/appropriateness ratings (subjective). 1v2=#9 architecture, 2v3=#13 compression, 1v3=bonus. Running the flat arm ourselves beats #9's between-study-vs-published comparison. Elad UNDECIDED, LEANING KEEP SEPARATE for now (both nearly ready; merge resets both) + cross-pollinate cheaply (add compressed arm to #9 or LIWC to #13). Flags: language (English keeps Client101 LIWC anchor vs internal Hebrew 3-arm), blinding+counterbalancing, fast=agents-only.

History

2026-07-02: 2026-07-02: 9<->13 3-arm merge recorded as OPEN option (Elad leaning keep separate); English-session constraint noted; both briefed to Karen.
2026-05-15: IRB cleared (Lavidor)

Related tasks

#53

Article #10 · Narrative Review (5-Axis)

LLM-Based Simulated Patients for Interpersonal Skills Training in Healthcare and Mental Health Education: A Five-Axis Narrative Review

Status UNDER REVIEW Methodology

Status note SUBMITTED to npj Digital Medicine 2026-07-07 (Collection: Transforming Medical Education through AI; Snapp ID 59b0d136; deadline was 24 Jul). JMIR MedEd declined ('too formative') 2026-07-06, transfer to i-JMR declined.

Authors Elyoseph, Z.*, Refoua, E.*, Holzman, Y., Haber, Y., Fargeon, M., Elyoseph, T., Linkovski, O., Yirmiya, K., Rafaeli, E.

Journal npj Digital Medicine

Data Manuscript at v18; Yuval preparing v19 for Sunday review meeting (close-out).

Folder study/articles/10.narrative review/ open

Last touched 2026-07-07

Attention ★ Action needed this week

Next step

npj technical check in progress -> peer review; decision ~4-10 wks

Notes

AUDIT FIX (rows 42, 59, 71): Updated from stale 'v15->v16 (Yuval doing final QA pass)' to 'v18 received comments -> Yuval preparing v19 for Sunday 2026-05-31 close-out'. This fix lands in 3 places (article card, Studies Needing Attention, What's Next timeline).

History

2026-07-07: 2026-07-07: Resubmitted to npj Digital Medicine collection after JMIR MedEd editorial decline; new cover letter (writer-agent, Zohar positioning); all 9 authors + MOST funding 0008682 declared
2026-07-06: 2026-07-06: SUBMITTED to JMIR Medical Education (submission #106300). Byline: Elyoseph (corresp+co-first), Holzman (co-first), Haber, Linkovski, Yirmiya, Rafaeli, Farjun, T. Elyoseph, Refoua. Status Unassigned (awaiting editor). Durable record: study/articles/10.narrative review/JMIR - medical education - final/SUBMISSION_NOTES_2026-07-06.md
2026-06-26: 2026-06-26: Yuval to finish by 2026-06-30 then meet Elad (#67).
2026-06-21: 2026-06-21: Table 1 row-by-row verification (one agent per row) + commented/corrected docx + ledger delivered to Yuval
2026-06-15: 2026-06-15: review almost completely done; will close + submit later this week (~2026-06-19)
2026-04-29: meeting / v16 target
2026-05-25: v18 comments received -> v19 in prep
2026-05-28: v19 prepared; final stages

Related tasks

#22 #23 #62 #67

Article #11 · Deep Persona

Deep Persona: A Three-Layered Psychological Architecture for Consistent and Credible Role-Playing Agents

Status UNDER REVIEW Methodology

Status note SUBMITTED 2026-04-30 by Rotem to TACL

Authors Dror, R. (1st), Refoua, E., Haber, Y., Elyoseph, Z., Solomon, A.

Journal TACL (target)

Data Theoretical/methodological paper on 3-layer psychological architecture (External -> Middle -> Internal).

Folder study/articles/11. Deep Persona/ open

Last touched 2026-06-21

Next step

Not in current focus; to revisit later (per Elad 2026-06-21).

Notes

TACL_Deep_Persona (13).pdf is LATEST draft (Apr 16). Rotem 1st author; Elad ~4th.

History

2026-04-30: submitted by Rotem

Article #12A · UP Pilot (Paper #1)

UP Pilot -- pre/post single module + AI experience moderator (Shaked-led feasibility paper)

Status DRAFTING Therapists

Status note UP -- M3 single-sim pilot framing SUPERSEDED 2026-07-02 (small-paper role moved to Article 18 lecture study)

Authors Seker, S. (1st), Refoua, E. (2nd), Linkovski, O., Elyoseph, Z., Yirmiya, K., Holzman, Y.

Journal Cognitive Therapy and Research (target)

Data Same dataset as 12B (N=71 canonical unique participants; BIU=32, Haifa=39). Pilot scope: pre/post single module + all bots + predictors with AI experience moderator. Optional qualitative add-on contingent on Tal (Task #18).

Folder study/articles/12. UP modules/ open

Last touched 2026-07-15

Attention ★ Action needed this week

Next step

SUPERSEDED by the 2026-07-02 Elad+Omer+Shaked call: the old 'small paper = M3 single-sim pilot from the course dataset + qualitative' plan is DROPPED (Omer: 'not really a pilot'; 06-28 decision = use ALL course data in the big paper). The program's small UP paper role now belongs to a NEW standalone lecture study = ARTICLE 18 (fresh data: UP-module lecture + one bot run for practicing therapists recruited via Facebook, pre/post, Zohar-IPT model). This 12A record is retained for audit continuity; the course-dataset within-subject analysis lives in 12B (N=76 / 711 LONG). Qualitative (Tal Eliyosef) folds into the big-paper story if used at all. No separate 12A submission is planned.

Notes

Split from Article 12 on 2026-05-25 per Shaked + Elad + Omer planning meeting. Shaked leads. Faster timeline than 12B. Same dataset; analyses must be split per paper to defend against salami-slicing critique. | shared with 12B; real at ~/Desktop/projects/UP simu/

History

2026-07-15: 2026-07-15 (menti-agent sync): corrected companion LONG obs to 711; confirmed no separate 12A submission (small-paper role = Article 18). No status change.
2026-07-06: 2026-07-02: Phone call (Elad+Omer+Shaked) DROPPED the M3-single-sim-pilot small-paper idea; the 'small paper' is replaced by a NEW standalone lecture study = Article 18 (different population/data/recruitment). 12A retained for audit; course-data analysis lives in 12B. (Source: UP simu research_log.md + PROJECT_TIMELINE.md; transcript 2026-07-02)
2026-07-02: 2026-07-02: next_step refreshed to v6 AUDIT-PASSED (N=76/702, not user-locked) during UP task reconciliation.
2026-06-29: 2026-06-29: cleaning being RE-DONE as v6 (v5 over-exclusion caught in Elad's code review); v6 plan FINAL awaiting approval. Propagated from UP simu 2026-06-29.
2026-06-09: 2026-06-08: UP cleaning pipeline 7-clean LOCKED v5 (N=74, BIU 33 + Haifa 41; 669 LONG obs; 3 codebooks built) -> Paper 1 analysis-ready. Propagated from UP simu 2026-06-09.
2026-05-25 PM: 2-paper SPLIT decision (split from prior Article 12)
2026-05-25 PM: Shaked confirmed as lead author

Related tasks

#18

Article #12B · UP Within-Subjects (Paper #2)

UP Within-Subjects -- WITH vs WITHOUT simulation x 3 modules x persistent Ethan (NPJ Digital Medicine target)

Status HARD DEADLINE 2026-07-31 Therapists

Status note UP BIG paper -- full study, Elad first author (RESHAPED 2026-07-02 call)

Authors Refoua, E. (1st), Seker, S. (2nd), Linkovski, O., Elyoseph, Z., Yirmiya, K., Holzman, Y.

Journal JMIR Medical Education (IF ~13.9; Nature Mental Health option left open) -- npj deadline DROPPED 2026-07-02

Data Same dataset as 12A. N=71 canonical unique participants (BIU=32, Haifa=39). 2026-05-25 auditable counts: BIU Q-Pre=53/Q1=77/Q2=48/Q3=60/Q4=49/Q5=46/Q6+Post=46. Haifa Q-Pre=55/Q1=40/Q2=30/Q3=56/Q4=29/Q5=44/Q6+Post=28.

Folder study/articles/12. UP modules/ open

Last touched 2026-07-15

Attention ★ Action needed this week

Next step

Full-study 'big paper' (Elad first author, target JMIR Medical Education; Nature Mental Health option open; npj deadline DROPPED 2026-07-02). Story = overall course pre/post improvement + sim-trained skills improved MORE than non-sim (per-module WITH-vs-WITHOUT). 2026-07-15: full Study-2 research plan WRITTEN (docs/STUDY2_FULL_RESEARCH_PLAN_2026-07-15) with a restructured analysis order (Elad 07-15): present per-module same-session matched contrasts FIRST (M3vsM2, M4vsM5/M6, M7vsM6 Haifa-only+flagged) then the pooled multilevel time x condition interaction as the registered confirmatory primary (module + participant + site random effects, one-tailed). Data cleaned via v6.1 pipeline: N=76 (BIU 34 + Haifa 42) / 711 LONG obs (v6.1 per-scale keep rule recovered +21 responses/+68 scale-obs vs v6; 0 scored values changed). STATUS (2026-07-15, Elad): (1) v6.1 cleaning is UNDER JOINT Elad+Shaked code review -> USER-LOCK (Shaked = second reviewer per readable-code plan; NOT a blocker, actively in progress). (2) PRE-REGISTRATION -- SCOPE RESOLVED (Elad 2026-07-15): the fresh pre-registration is to be PREPARED FROM SCRATCH FOR 12B ('tzarich lehachin le-12B') -- this is the active upstream pre-reg deliverable. There is NO finalized/approved/registered pre-reg for 12B; prior EARLY WORKING DRAFTS (OSF-style preregistration_draft_v4.0 and older v3.4) are SUPERSEDED starting points, NEITHER finalized NEITHER ever registered on OSF. The fresh pre-reg is IN PREPARATION by Elad/team; the OSF-registration click stays pending Elad (iron rule #11). (Article 18 already has its own working draft -- handled separately.) (3) Barlow co-authorship CONFIRMED to be approached at the END, after a draft exists. THEN: finalize the fresh pre-analysis registration with honest data-already-collected disclosure + analysis code (r-coder) -> run -> manuscript (writer).

Notes

Split from Article 12 on 2026-05-25 per Shaked + Elad + Omer planning meeting. Elad leads. Major novelty: persistent simulation (Ethan across 3 modules) + within-subject WITH-vs-WITHOUT simulation comparison + AI-experience moderator. Tasks #14, #15, #18, #21 all block this paper; Task #20 is the submission. | shared with 12A; real at ~/Desktop/projects/UP simu/

History

2026-07-15: 2026-07-15: SCOPE flag resolved (Elad) -- fresh pre-reg to be PREPARED FROM SCRATCH for 12B; this is the active upstream pre-reg deliverable. Removed 'SCOPE=CONFIRM WITH ELAD'.
2026-07-15: 2026-07-15: Elad update — (a) the UP study/design is built from ZOHAR's IPT protocol (Zohar sent it). (b) Shaked has FINISHED reviewing the cleaning code; the FINAL JOINT Elad+Shaked review (-> USER-LOCK of v6.1, N=76/711) is scheduled for TOMORROW, Thu 2026-07-16.
2026-07-15: 2026-07-15: correction (Elad) -- NO finalized/approved/registered pre-reg exists for 12B; it is being REDONE FROM SCRATCH; v4.0/v3.4 = superseded early drafts, not near-ready. Scope of the fresh redo (12B/Art.18/both) = confirm with Elad.
2026-07-15: 2026-07-15: Absorbed 3 Elad UP status updates -- v6.1 USER-LOCK now under joint Elad+Shaked code review (in progress, not blocked); open decisions + pre-reg in prep by Elad/team (OSF click still pending Elad); Barlow deferred-to-post-draft CONFIRMED.
2026-07-15: 2026-07-15 (menti-agent sync from UP simu): corrected LONG obs 702->711 (v6.1 per-scale keep rule); recorded Study-2 full research plan written + analysis order restructured to per-module-matched-contrasts-first (Elad 07-15). No status change (stays in_preparation); no inferential analysis run yet.
2026-07-06: 2026-07-02: Phone call (Elad+Omer+Shaked) RESHAPED the papers. This 12B = the full-study BIG paper. Target journal npj DM special issue -> JMIR Medical Education (IF ~13.9; Nature Mental Health option open). npj 31-Jul-2026 HARD deadline DROPPED (unattainable). Predictors/anthropomorphism paper DROPPED. Barlow deferred post-draft. Elad = first author. Cleaning code being rebuilt for readability; Shaked signs off as second reviewer. (Source: UP simu docs/research_log.md top + PROJECT_TIMELINE.md; transcript tl_omer_shaked_call_2026-07-02.txt)
2026-07-02: 2026-07-02: next_step refreshed to v6 AUDIT-PASSED (N=76/702, not user-locked; prereg v4.0 drafted not registered) during UP task reconciliation.
2026-06-29: 2026-06-29: npj DM deadline confirmed 31 Jul 2026 (corrected from a stale 24 Jul online reading).
2026-06-29: 2026-06-29: cleaning being RE-DONE as v6 (v5 over-exclusion); prereg v3.4 found NEVER registered on OSF; v4.0 drafted for Paper 2, sent to UP group, not registered. Propagated from UP simu 2026-06-29.
2026-06-09: npj DM deadline confirmed 2026-07-31 per the CFP site (Elad 2026-06-09); 07-24 / 06-26 variants superseded.
2026-06-09: 2026-06-08: UP cleaning pipeline 7-clean LOCKED v5 (N=74; 669 LONG obs; 3 codebooks: GUIDE/empirical/meaning-map) -> Paper 2 analysis-ready. Propagated from UP simu 2026-06-09.
2026-05-15: data collection done
2026-05-17: pre-reg v3.4 to UP team
2026-05-25 evening: tracking overhaul (N=71 canonical)
2026-05-25 PM: 2-paper SPLIT decision
2026-05-26: Joy Haifa in-person session

Related tasks

#14 #15 #16 #17 #19 #20 #74

Article #13 · Prompt Compression

Prompt Compression Study — empirical validation of CPE via blinded expert rating

Status RUNNING Methodology

Status note IN PREP — ACTIVELY RUNNING (expert platform live; Yuval recruiting raters)

Authors Refoua, E. (with Holzman Y., Rafaeli E. supervising; full list TBD)

Journal JMIR Med Educ (primary) / JMIR Mental Health / CHB:AH (backups)

Data Pre-registered within-subjects blinded evaluation. 10 pairs x 14 items x 2 conversations. 20 Hebrew therapeutic sim conversations.

Folder study/articles/13. Prompt Compression/ open

Last touched 2026-07-02

Next step

Rating platform live; within-subjects blinded eval (10 pairs x 14 items x 2 conversations). WATCH: 9<->13 merge option under consideration (see notes) - Elad leaning keep separate for now.

Notes

IRB Amendment 1 covered by Lavidor email 2026-05-15. Rating platform is LIVE. OPEN STRATEGIC OPTION (Elad 2026-07-01, cross-linked to #9, NOT a decision): merge #9+#13 into one 3-arm within-subjects study - (1) flat/Client101 baseline, (2) full CPE ~33K, (3) compressed CPE ~3K; 10 therapists each run all 3 (+optionally AI agents); measured by BOTH LIWC (objective) AND clinician richness/appropriateness ratings (subjective). 2v3=#13 compression, 1v2=#9 architecture, 1v3=bonus. Elad UNDECIDED, LEANING KEEP SEPARATE for now (both nearly ready - #13 platform live, #9 protocol-ready; merge resets both) + cross-pollinate cheaply (add LIWC to #13 or a compressed arm to #9). Flags: language (English keeps Client101 LIWC anchor vs internal Hebrew 3-arm), blinding+counterbalancing, fast=agents-only.

History

2026-07-02: 2026-07-02: 9<->13 3-arm merge recorded as OPEN option (Elad leaning keep separate); briefed to Karen.
2026-04-16: last activity

Related tasks

#55 #56

Article #14 · Clalit / Dalia

Clalit / Dalia Study — bot for parent groups in Clalit Rishon LeZion HMO

Status PROTOCOL Parent

Status note PROTOCOL SUBMITTED — Dalia in ethics course

Authors Dalia (PI Clalit), Refoua, E., Holzman, Y. (operationally)

Data Protocol submitted to Clalit ethics; Dalia completing required ethics course.

Folder study/articles/14.clalit/ open

Last touched 2026-07-15

Verify ⚠ Verification needed

Next step

ADVANCE the study. Blocker = the contract/undertaking: fix the NAME on it AND add Elad as a signatory. Next action = Elad talks to Omer (Linkovski) about this. (Prior gate: Zohar+Eshkol consult on IP + who-signs-as-PI; PI-name reconcile — Dalia Kelly Benacot per protocol vs now driven via Omer+Karen? confirm w/ Elad.)

Notes

Per audit row 46: VERIFY NEEDED — first MENTI deployment in an Israeli HMO. Could 'catapult us' (Elad's words).

History

2026-07-15: 2026-07-15: Elad update — advance A14. The 'signatures' issue = a move to fix the contract NAME + Elad wants to be a signatory. Next action = Elad -> Omer (Linkovski). Reconcile PI (Dalia Kelly Benacot on protocol vs Omer+Karen-driven now) flagged.
2026-06-28: 2026-06-28: Zohar briefed in depth via WhatsApp + sent נספח א' + track-changes undertaking; open PI question (Elad/Karen/Zohar); awaiting Zohar+Eshkol consult before replying to Tziki. (Earlier 06-26 briefing deleted by Elad — wrong population/too shallow — corrected to parents-focused.)
2026-06-26: 2026-06-26: Elad confirms Clalit IP/undertaking needs consult with BOTH Zohar AND Eshkol (was 'Zohar and possibly Eshkol') before replying to Tziki.
2026-06-25: 2026-06-25: Elad wants to consult Zohar (+ maybe Eshkol) on the IP/undertaking before sending to Tziki.
2026-06-25: 2026-06-25: submission package assembled; undertaking IP clause revised IN-PLACE in track-changes for legal-counsel re-approval; PI surname unified to בנאקוט; clinical-manager placeholder removed.
2026-06-15: 2026-06-15: Clalit sent an email (now in hand); proposal to be updated accordingly by Elad (task #50).
2026-06-15: 2026-06-15: new email from Clalit received; proposal to be updated accordingly (Elad)
2026-06-08: Meeting 2026-06-08: corrected protocol discussed; Elad to send/sent to Dalia + Barak (Cesura security, no direct clinical risk, Claude-based CPE).
2026-06-08: Applied Elad's inline 'koud' comments to the Clalit protocol -> corrected docx (2026-06-08); original preserved; parallel-diff confirms only the 3 corrections changed.
2026-04-29: protocol submitted

Related tasks

#50

Article #15 · Rina

Rina Counseling-Student Bot Evaluation (SDT / needs-supportive dialogue)

Status QUESTIONNAIRE LIVE Therapists

Status note v29 live (+Course Eval + reflection; de-counseled; BotQ 5 / Learning 7 / SUS 6); sent to Inbal; awaiting Thu 2026-06-18 run

Authors Refoua, E., Assor (Prof. Avi Assor, BGU)

Data Qualtrics SV_734qjklvoJkl4to LIVE at v19. 7 blocks, 23 substantive items. Hebrew RTL. Target N=40-80.

Folder study/articles/15. Rina/ open

Last touched 2026-07-15

Attention ★ Action needed this week

Next step

Two live surveys (NEVER pool). INTERVENTION SV_734qjklvoJkl4to: 45 real = 37 completed + 8 partial (raw 55 minus 10 dev/test recorded <=2026-06-17, still never wiped, exclude by date at analysis); 06-18 run = 13 completed. CONTROL SV_3CQw2i9BhfgMuAm (non-course; Gordon/Shaanan/David Yellin): 4 real = 1 completed + 3 partial -- control recruitment is the bottleneck. Both surveys active. NEXT: push control-group recruitment (Inbal status + control-link push sent 2026-07-12); dev/test wipe pending Elad; QID19 program-option decision (Elad).

Notes

AUDIT FIX (row 47): Updated from stale 'v15 published' to 'v19 published'. Version history milestones added: v18 = Inbal edits applied (2026-05-18 mid-day), v19 = Article# removal + SUS/WAI proposal sent (2026-05-18 evening). Pre-v12 history: v11 gender-inclusive, v10 Look-and-Feel, v8 4-fix, v6 consent.

History

2026-07-15: 2026-07-15: Elad update — OPEN COORDINATION ITEM: a meeting with Inbal Cohen (ענבל, David Yellin field coordinator) needs to be scheduled about Rina's bot. Elad to set it up.
2026-07-12: 2026-07-12: response-status snapshot (metadata only) -- intervention 45 real (37 completed/8 partial), control 4 real (1 completed/3 partial); status + control-link push sent to Inbal (each once, signed).
2026-07-03: 2026-07-03: fresh pull - INTERVENTION 30 real completions since 06-18 (13 on 18.6 + 11 on 23.6 + trickle; 17 pre-06-18 dev/test rows to exclude); CONTROL stuck at 4 (all 23.6). Sent Inbal a WhatsApp ONCE (msg 3EB0D7F8..., 00:14 IL, signed Claude) asking to redistribute the control link at Gordon/Shaanan
2026-07-02: [dashboard] Task #65 status -> closed (by אלעד)
2026-06-15: 2026-06-15: removed recommend_other (QID25) from Recommendation block, published v27 (non-destructive; matrix kept)
2026-06-15: 2026-06-15: applied Yuval's 2 edits to live, published v25 (consent IRB parenthetical removed + bot_quality_1 feminized)
2026-06-15: 2026-06-15: v23 done + with Yuval (send tomorrow after review); Cesura Learning Space to prepare; final sim run Thu 2026-06-18 with questionnaire
2026-05-28: SUS + WAI-TECH-SF both APPROVED by Inbal; back-translation review with Yuval is the gate before upload
2026-05-17: built end-to-end
2026-05-17 PM: consent fix v6
2026-05-17 PM: 4-fix v8
2026-05-17 PM late: Look-and-Feel v9
2026-05-17 evening: gender-inclusive v11
2026-05-18: v12 restructure (Perceived Learning)
2026-05-18 mid-day: v18 Inbal edits applied
2026-05-18 evening: v19 Article# removed + SUS/WAI proposal sent

Related tasks

#5 #6 #7 #8 #9 #10 #31 #64 #65 #77

Article #16 · Menti Zimbardo Style

Menti Zimbardo Style (stub)

Status STUB Methodology

Status note STUB — folder placeholder, no codebook yet

Authors TBD

Data Folder exists at study/articles/16. Menti zimbardo style/; no codebook, no protocol, no scoping document yet.

Folder study/articles/16. Menti zimbardo style open

Last touched 2026-07-02

Next step

Stub - parked; scope when Elad raises it (reviewed 2026-07-02)

Notes

Registered for article-count consistency across project memory + dashboard. Folder predates 2026-05-28 registration; content TBD.

History

2026-05-28: registered as stub for count consistency

Article #17 · Hili Pilot — MentiTherapist

MentiTherapist for Parent-Work (Hili Pilot)

Status DATA COLLECTION DONE TherapistsParent

Status note PILOT RAN + analyzed 2026-06-07 (N=29); both primaries supported; preliminary report shared

Authors Elad Refoua + Dr. Hili Kochavi + Yuval Holzman

Data PILOT RAN Sun 2026-06-07 ~12:00 IL @ MCLA Ramat Gan (MA Educational Psychology trainees, year 1; Lia parent-sim; within-subjects pre-post single arm). PRE 32 / POST 31 / LINKED N=29 (last-4-ID). Analyzed same day. BOTH primaries supported: H1 Scale C SE-for-guiding-parents 5.11->5.33 d_z=0.55 p_holm=.013; H2 Scale B apprehension 2.88->2.63 d_z=-0.47 p_holm=.018. Challenge subscale d_z=.59 > Foundational d_z=.38 (Yoav-consistent). Acceptability composite 5.57/7; recommend 22/29=76% (p=.004). RFQ-8 flat (H7 null, uninformative). Numbers in Analysis_Pilot_2026-06-07/stats.json.

Folder study/articles/17. Hili - parent simulation pilot open

Last touched 2026-06-15

Attention ★ Action needed this week

Next step

v3 manuscript AUDIT-PASSED + Word sent to Elad (USER-LOCK pending, task #63). NEXT HUMAN STEP: Zohar works on the v3 manuscript this week (week of 2026-06-15), alongside Elad's line-by-line review. 3 open decisions: Scale B status (co-primary vs alpha .39/.62) / length (~6300 vs Brief-Report 4-5k) / author lists. Target: educational/school-psych Q1 Brief Report (School Psychology Review; alt Training and Education in Professional Psychology).

Notes

V2 design migration done 2026-05-29 (Scale A dropped, Scale B 1-5 -> 1-7, sim_feedback_3 to המדריכה, open_anxieties + open_else dropped, open_positive + open_negative added Yoav-style). RFQ-8 (Fonagy et al. 2016) added as H7 SECONDARY mechanism. 25 QC rounds, 5/5 final streak achieved. LIVE PRE 5 blocks 17Q + LIVE POST 6 blocks 14Q. Timing 10+25+10=45 min. Site v15.12 at https://elad-refoua.github.io/article17-hili-pilot/.

History

2026-06-15: 2026-06-15: Zohar to work on the v3 manuscript this week (week of 2026-06-15)
2026-06-09: 2026-06-09: v3 manuscript AUDIT-PASSED + Word built + sent to Elad
2026-06-08: Meeting 2026-06-08: Hili/Lia proposed moving the study to 2nd-year MA students working with parents in the field (path to a co-authored journal paper); budget = cost of ~+50 participants; decision pending.
2026-06-07: 2026-06-07: PILOT RAN @ MCLA; N=29 linked; both primaries supported; preliminary report shared with team
2026-05-30: 2026-05-29 to 2026-05-30: V2+RFQ-8 design migration applied, 25 QC rounds run sequentially, 5 consecutive clean streak achieved at round 25. Package delivered to Yuval Holzman via WhatsApp 2026-05-29 21:08 IL (4 docx + xlsx files, Hebrew message, signed Claude מטעם אלעד). Pilot date corrected 2026-06-08 -> 2026-06-07 (Sunday).
2026-05-28: 2026-05-28 Elad: article spans therapists (trainees) + parent (training content) tracks
2026-05-27: intro meeting (Hili + Elad); folder opened
2026-05-27: Hili emails Raviv-Bar-Tal scale + Kochavi apprehension scale
2026-05-28: protocol v5 APPROVED by Elad; Scale C v1 drafted

Related tasks

#26 #27 #28 #29 #37 #40 #41 #42 #43 #44 #46 #47 #63

Article #18 · UP Lecture Study

UP Lecture Study -- UP-module lecture + AI simulation for practicing therapists (pre/post, Facebook recruitment, Zohar-IPT model)

Status PLANNING Therapists

Status note NEW small paper from the 2026-07-02 UP call -- standalone lecture study (replaces the dropped M3-pilot)

Authors Refoua, Seker, Linkovski (order TBD); MENTI standard co-authors TBD

Journal TBD (Zohar's IPT-evenings papers are the model)

Data Not started -- design phase. Recruitment via Facebook (Shaked's 'shlishi ha-melachuti'), cap ~100 therapists; new dedicated Cesura workspace to be opened.

Folder study/articles/18. UP lecture study/ open

Last touched 2026-07-15

Next step

NEW standalone 'lecture study' decided in the 2026-07-02 Elad+Omer+Shaked phone call, replacing the dropped M3-pilot as the program's small UP paper. Design (modeled on Zohar Elyoseph's IPT evenings, each of which became its own paper): an evening / Friday-morning LECTURE on ONE UP module + an AI simulation run once, for PRACTICING therapists; pre/post questionnaires. Recruitment via Facebook (Shaked's 'ha-shlishi ha-melachuti' page), cap ~100. This is FRESH data with a DIFFERENT population -- NOT a re-analysis of the BIU/Haifa course dataset (that is Article 12B). ROLES: Shaked -- owns pushing the initiative, prepares the DEDICATED questionnaire + opens a NEW Cesura workspace, second-reviewer sign-off on cleaning code; Omer -- prepares and DELIVERS the lecture; Elad -- operational (build + publish the flyer) + consults Zohar on the study design + reviewer lessons from Zohar's IPT papers (meets him soon). Considering one module (Shaked's lean preference) vs all seven. Existing UP post-questionnaires can be reused. PRE-REGISTRATION (SCOPE RESOLVED, Elad 2026-07-15): Article 18 ALREADY HAS a working pre-reg draft -- an EARLY WORKING AsPredicted DRAFT from 2026-07-14 (docs/STUDY1_M3_LECTURE_PREREG_ASPREDICTED_2026-07-14) with open decisions still listed. It is NOT finalized and NOT registered on AsPredicted -- but a draft EXISTS. Article 18 is NOT the record being prepared-from-scratch (that is 12B). Do NOT treat this draft as approved/registered; the AsPredicted-registration click stays pending Elad (iron rule #11). NEXT: Elad to talk to Zohar re: design + reviewer lessons; Shaked to build questionnaire + Cesura workspace + push recruitment; Elad flyer.

Notes

Cross-links: Article 12 (UP; shares bots/instruments but SEPARATE data + population), Article 8 (Zohar's IPT-evenings publication model). Home context = UP simu external project (C:/Users/user/Desktop/projects/UP simu/). menti_track = therapists (Rafaeli+Linkovski).

History

2026-07-15: 2026-07-15: SCOPE flag resolved (Elad) -- Art.18 draft EXISTS (07-14 AsPredicted, not finalized/registered); it is NOT the prepared-from-scratch record (that is 12B). Dropped 'SCOPE=CONFIRM WITH ELAD'.
2026-07-15: 2026-07-15: correction (Elad) -- the 07-14 AsPredicted item is an EARLY WORKING DRAFT, not finalized/registered; no finalized pre-reg exists for Art.18. Whether the 'redo from scratch' covers Art.18 (vs 12B only) = confirm with Elad.
2026-07-06: created via ops.py article add

Related tasks

#75