claude

bchanot/claude

Fork 0

Commit Graph

Author	SHA1	Message	Date
bastien	11b0c6bb8e	darwin: add test prompts for 5 skills (baseline pass) Skills covered: close, graphify, harden, profile, prune-memory. Used by /darwin-skill dim 8 effect testing. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-12 17:50:06 +02:00
bastien	0fe6153ead	fix(prune-memory): STEP 4 verify — prefix mapping bug (TDD RED→GREEN) First end-to-end run of /prune-memory on real .claude/memory/ surfaced a broken verify script: Old: `prefix=$(basename "$f" .md \| tr a-z A-Z \| cut -c1-3)` derived the prefix from the filename's first 3 letters → produced DEC / LEA / BLO. Actual prefixes are BDR / LRN / BLK. The grep then matched zero entries, no MISSING/ORPHAN was ever reported, and the script printed its "OK if blank" footer regardless of real state. False clean signal. Fixed: hard-mapped filename → prefix via `declare -A PREFIX_MAP`. Verified against current registries — 14 BDR + 16 LRN + 2 BLK + 1 EVAL entries all index-consistent, no false negatives. Added EVAL prefix to the map (evals.md was missing from the loop in v1). Footer line clarified to `(blank above = OK)`. `wc -l` excludes `.original.md` backups from the output. Note: caveat in skill body said "v1 ships without baseline TDD test — STEP 2 approval gate is the safety net". First real test caught a verify bug that bypassed STEP 2 entirely. Lesson: STEP 4 is its own safety net and needs its own test. Co-Authored-By: Claude <noreply@anthropic.com>	2026-05-11 16:27:31 +02:00
bastien	fd619f5f9a	feat(skills): add /prune-memory — curate .claude/memory/ registries New personal skill to maintain memory registry hygiene. Gap identified between existing tools: - /caveman:compress — text-compresses one file, no curation - /close — appends new entries end-of-session, doesn't prune - /prune-memory (new) — audits, classifies, applies user-approved cleanup Operations: - Mark obsolete entries `status: superseded by <ID>` or `status: deprecated` (no hard delete — append-only per CLAUDE.md memory rule). - Merge similar entries (new ID, sources marked superseded). - Caveman-compress bloated prose-heavy entries inline. - Repair Index drift (missing rows, orphaned rows). Workflow: STEP 0 precheck (refuses dirty working tree, git = backup) → STEP 1 audit (A obsolete / B similar / C bloated / D drift) → STEP 2 plan + mandatory user approval → STEP 3 apply safe→destructive → STEP 4 verify Index sanity + line-count report. Follows superpowers:writing-skills CSO conventions: "Use when..." trigger description (under 1024-char spec), Quick Reference table, Common Mistakes table, Failure Paths table. v1 ships without baseline TDD test (noted in skill body); STEP 2 approval gate is the safety net. Co-Authored-By: Claude <noreply@anthropic.com>	2026-05-11 16:24:16 +02:00

Author

SHA1

Message

Date

bastien

11b0c6bb8e

darwin: add test prompts for 5 skills (baseline pass)

Skills covered: close, graphify, harden, profile, prune-memory.
Used by /darwin-skill dim 8 effect testing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-12 17:50:06 +02:00

bastien

0fe6153ead

fix(prune-memory): STEP 4 verify — prefix mapping bug (TDD RED→GREEN)

First end-to-end run of /prune-memory on real .claude/memory/ surfaced
a broken verify script:

Old: `prefix=$(basename "$f" .md | tr a-z A-Z | cut -c1-3)` derived
the prefix from the filename's first 3 letters → produced DEC / LEA /
BLO. Actual prefixes are BDR / LRN / BLK. The grep then matched zero
entries, no MISSING/ORPHAN was ever reported, and the script printed
its "OK if blank" footer regardless of real state. False clean signal.

Fixed: hard-mapped filename → prefix via `declare -A PREFIX_MAP`.
Verified against current registries — 14 BDR + 16 LRN + 2 BLK + 1 EVAL
entries all index-consistent, no false negatives.

Added EVAL prefix to the map (evals.md was missing from the loop in
v1). Footer line clarified to `(blank above = OK)`. `wc -l` excludes
`.original.md` backups from the output.

Note: caveat in skill body said "v1 ships without baseline TDD test —
STEP 2 approval gate is the safety net". First real test caught a
verify bug that bypassed STEP 2 entirely. Lesson: STEP 4 is its own
safety net and needs its own test.

Co-Authored-By: Claude <noreply@anthropic.com>

2026-05-11 16:27:31 +02:00

bastien

fd619f5f9a

feat(skills): add /prune-memory — curate .claude/memory/ registries

New personal skill to maintain memory registry hygiene. Gap identified
between existing tools:
- /caveman:compress — text-compresses one file, no curation
- /close — appends new entries end-of-session, doesn't prune
- /prune-memory (new) — audits, classifies, applies user-approved cleanup

Operations:
- Mark obsolete entries `status: superseded by <ID>` or `status: deprecated`
  (no hard delete — append-only per CLAUDE.md memory rule).
- Merge similar entries (new ID, sources marked superseded).
- Caveman-compress bloated prose-heavy entries inline.
- Repair Index drift (missing rows, orphaned rows).

Workflow: STEP 0 precheck (refuses dirty working tree, git = backup)
→ STEP 1 audit (A obsolete / B similar / C bloated / D drift)
→ STEP 2 plan + mandatory user approval
→ STEP 3 apply safe→destructive
→ STEP 4 verify Index sanity + line-count report.

Follows superpowers:writing-skills CSO conventions: "Use when..." trigger
description (under 1024-char spec), Quick Reference table, Common
Mistakes table, Failure Paths table. v1 ships without baseline TDD test
(noted in skill body); STEP 2 approval gate is the safety net.

Co-Authored-By: Claude <noreply@anthropic.com>

2026-05-11 16:24:16 +02:00

3 Commits