agentleFS
Sign inStart free
michelgroletmichelgrolet/people-memory-mcp/skills/maintain-people-memory/SKILL.md

maintain-people-memory

Skill1 starsChanged 3 months agoMITView on GitHub

Audit and maintain a People Memory graph: find probable duplicate people or organizations, orphan records, conflicting facts, missing identifiers, stale relationships, invalid dates, import quality issues, or oversized archives; propose merges and cleanup, back up first, and apply only user-approved destructive changes. Use for doctor, audit, cleanup, dedupe, merge, stale contacts, graph health, data quality, restore, or repair requests.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start read-only. Show evidence for each proposed change.

    line 13
  2. Distinctive:

    Use guarded SQL to find exact normalized duplicate names and organizations.

    line 18
  3. Distinctive:

    Search similar names only as candidates, never as automatic duplicates.

    line 19
  4. Very specific:

    Call stale_contacts separately.

    line 22
  5. Very specific:

    Check deleted_records size and backup age.

    line 23
  6. Distinctive:

    Ask before merges, deletes, archive purges, provenance changes, or bulk updates.

    line 29
  7. Distinctive:

    When merging people, choose a survivor with the strongest identifiers and most user-stated data.

    line 32

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Maintain People Memory

Start read-only. Show evidence for each proposed change.

Audit

  1. Call graph_status.
  2. Use guarded SQL to find exact normalized duplicate names and organizations.
  3. Search similar names only as candidates, never as automatic duplicates.
  4. Find people without identifiers, affiliations pointing to deleted history mistakes, facts without useful provenance, invalid dates, self-edges, and conflicting current affiliations.
  5. Call stale_contacts separately. Staleness is a relationship prompt, not a data defect.
  6. Check deleted_records size and backup age.

Group findings into safe automatic fixes, identity decisions, destructive cleanup, and suggestions.

Change gate

Ask before merges, deletes, archive purges, provenance changes, or bulk updates. Back up the database before approved destructive work. Use a transaction and verify row counts before commit.

When merging people, choose a survivor with the strongest identifiers and most user-stated data. Move identifiers, affiliations, facts, interactions, and both sides of relationships. Resolve unique key conflicts explicitly. Archive both pre-merge records. Delete the duplicate last.

Verify

Re-run the audit, read the survivor record, and verify the UI search. Report exactly what changed and what remains unresolved. Do not expose third-party data beyond the minimum needed for the owner's decision.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free