4 Commits

Include renames

Author	SHA1	Message	Date
Magnus Hedemarkandfactory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>	47cd4d430a	chore(dsm5): add Available Scripts table and Prerequisites/Limitations ... Document lookup.py in an Available Scripts table with invocation and run-when guidance tied to the routing workflow; add Prerequisites and Limitations reflecting the stdlib-only local-library search. Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>	2026-08-22 23:14:30 -04:00
Magnus HedemarkandGitHub	5d101007ef	fix(skill): SkillOpt 3-epoch optimization — dsm5 navigation, decisions, and answer patterns (#281) ... Greenfield post-publication SkillOpt run (epoch 0 = 3eb7bd4) on dsm5. Epoch 1 (prominence/navigation, 4 edits): - Promote scripts/lookup.py into workflow step 3 for unknown-condition routing - Anchor split-chapter index-first pattern at the routing table - Cross-reference Completion criteria from workflow step 9 - Add read-scope guard to step 4 (index + specific part, no whole chapters) Epoch 2 (decision intelligence, 3 edits): - Answer-first rule: provisional answer with marked unknowns, then high-yield follow-ups - Third-party rule: asker may not be subject; never diagnose third parties - Code-verification guard: cite codes/specifiers/prevalence only from read files Epoch 3 (pattern expansion, 2 edits): - Add Answer shape section: six-part structure calibrating output depth - Add existing-diagnosis handling to Audience adaptation (explain, don't re-derive) Validation: 21 held-out runs (baseline+candidate), zero regressions; final installed-artifact regression passed. Description unchanged; catalogs unaffected.	2026-08-05 13:00:09 -04:00
3 people	c312b36166	fix(skill): split oversized dsm5 references into index + parts (#278) ... Agent harnesses truncate file reads around ~60k characters, so the largest dsm5 reference files (up to 132k chars) were being cut off mid-file (reported: "The neurodevelopmental file was truncated"). - Split 15 reference files over 50k chars into a small index (original filename preserved, so all existing links keep resolving) plus part files of <= ~40k chars each, organized by disorder group - Updated SKILL.md routing rows to point at indexes and read the part for the condition; added large-file handling guidance - Updated dsm5/README.md What You Get table; documented the size convention in 00-overview-and-method.md (Maintaining this library) - Verified: no reference file exceeds 50k chars (66 files), all 466 relative links resolve, validators pass, lookup.py lists all parts Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>	2026-08-04 22:10:15 -04:00
3 people	d346970bf8	feat(skill): add dsm5 — evidence-based companion to the DSM-5-TR (#276) ... Adds the dsm5 skill: an evidence-based conversational expert grounded in the DSM-5-TR (American Psychiatric Association, 2022) for clinicians, practitioners, patients, and family members. - SKILL.md: safety-first conversation workflow (triage -> clarify -> route -> compare criteria -> differentials -> calibrated conclusion), reference routing table, crisis protocol, audience adaptation - references/: 28 files — foundation (00-02), all 22 DSM-5-TR diagnostic classes (10-31), Part III measures/culture/AMPD/conditions-for-further- study (32-33), and cross-cutting differentials (40). Criteria are paraphrased with exact counts, durations, specifiers, and ICD-10-CM codes, plus per-disorder clinician and patient/family conversation guides - scripts/lookup.py: stdlib keyword search across the reference library (--json/--list/--max/-q) - evals/evals.json: 9 output-quality cases (schema v1) - README.md: human-facing overview, install notes, and APA attribution - Catalog entries and generated artifacts (llms.txt, marketplace plugins) regenerated; all repo validators pass Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>	2026-08-04 21:30:11 -04:00