Review and maintain EC memories. List all memories, find duplicates, prune stale entries, and organize by type. Use when memories need cleanup, when user says "audit memories", "clean up EC", "review what's stored", or periodically for hygiene.
Audits and organizes stored memories by identifying duplicates, removing stale entries, and categorizing content for optimal recall.
npx claudepluginhub merewhiplash/engram-cogitatorThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Review, clean, and organize stored memories.
Announce: "I'm using the audit skill to review EC memories."
List All → Analyze → Present → Clean → Verify
Pull everything by type:
ec_list:
limit: 100
type: decision
ec_list:
limit: 100
type: learning
ec_list:
limit: 100
type: pattern
ec_list:
limit: 100
type: config
Also check for invalidated entries:
ec_list:
limit: 50
include_invalid: true
Organize memories by their area tag to see coverage:
## EC Memory Audit
**Total Memories:** N (D decisions, L learnings, P patterns, C config)
**Invalidated:** N
### By Area
| Area | Decisions | Learnings | Patterns |
|------|-----------|-----------|----------|
| auth | 3 | 2 | 1 |
| api | 1 | 4 | 2 |
| ... | ... | ... | ... |
### Issues Found
- **Duplicates (N):** [list IDs]
- **Potentially stale (N):** [list with reason]
- **Vague entries (N):** [list IDs]
Use AskUserQuestion for each category:
{
"questions": [{
"question": "How should I handle the N duplicate memories?",
"header": "Duplicates",
"options": [
{ "label": "Merge", "description": "Keep best version, invalidate others" },
{ "label": "Keep all", "description": "They might have nuance" },
{ "label": "Review each", "description": "Show me one by one" }
],
"multiSelect": false
}]
}
When removing a memory, use ec_invalidate with optional superseding:
ec_invalidate:
id: <old_memory_id>
superseded_by: <new_memory_id> # Optional: link to replacement
ec_addec_invalidate, pointing to new IDAfter cleanup, show the delta:
## Cleanup Complete
**Before:** N memories
**After:** M memories
**Invalidated:** X entries
### Changes
- Merged N duplicates
- Invalidated M stale entries
- Recategorized P entries
| Check | Action |
|---|---|
| Duplicate content | Merge or invalidate |
| References deleted code | Invalidate |
| Too vague to act on | Rewrite or invalidate |
| Wrong type | Create new with correct type, invalidate old |
| Missing area tag | Update via new entry |
Stop and ask if:
Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
Search, retrieve, and install Agent Skills from the prompts.chat registry using MCP tools. Use when the user asks to find skills, browse skill catalogs, install a skill for Claude, or extend Claude's capabilities with reusable AI agent components.
This skill should be used when the user wants to "create a skill", "add a skill to plugin", "write a new skill", "improve skill description", "organize skill content", or needs guidance on skill structure, progressive disclosure, or skill development best practices for Claude Code plugins.