Menu
Two-tier memory management: write, update, prune, and search memories with semantic deduplication. Maintains MEMORY.md index under 200 lines.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Produces a numbered Architecture Decision Record with context analysis, route comparison, decision rationale, consequences, and propagation checklist. Use when: 'write an ADR', 'architectural decision', 'record a decision', 'ADR for'.
Audits CLAUDE.md files for conflicts, stale rules, missing sections, and cross-file coherence. Produces a health report with auto-fix suggestions. Use when: 'check claude.md', 'audit CLAUDE.md', 'fix claude.md', 'claude md health'.
Structured codebase intelligence: directory tree with roles, entry points, dependency graph, test coverage map, and annotated 'here be dragons' warnings.
Runs a convergence gate: counts dirty files, sessions since last convergence, open items, and produces a triage report with RED/YELLOW/GREEN status. Use when: 'converge', 'convergence check', 'drift check', 'am I drifting'.
Bootstrap project structure from proven templates: phases, tracks, decisions, artifacts, commissions directories with README stubs and initial ADR-000 template.
Produces a scout document: identifies a tension, generates 3-5 distinct routes with tradeoffs, synthesizes a recommendation, and proposes a first action. Use when: 'scout this', 'strategic scout', 'explore options for', 'what are the routes'.
| name | memory-curator |
| description | Two-tier memory management: write, update, prune, and search memories with semantic deduplication. Maintains MEMORY.md index under 200 lines. |
| triggers | ["manage memories","curate memory","prune memories","memory health","organize memories"] |
| version | 1.0.0 |
| model | sonnet |
Manages the persistent memory system — writes, updates, prunes, and deduplicates memories while keeping the MEMORY.md index lean and navigable.
Read MEMORY.md index and all referenced memory files in the memory directory. Build an inventory of:
Detect semantically duplicate memories — same fact stored with different wording. Criteria:
Identify memories referencing things that no longer exist:
Combine related memories that cover the same topic:
Remove memories that are:
Rebuild MEMORY.md:
Output summary:
