agentleFS
jin2004-cmdjin2004-cmd/codex-practical-skills/personal-secretary-memory/SKILL.md

personal-secretary-memory

Skill1 starsChanged 36 days agoMITView on GitHub

Maintain or recover a user's dated personal-assistant memory and prepare a cross-assistant handoff. Use only for explicit memory updates, continuity recovery, or handoff work; ordinary career, video, website, or life tasks alone do not trigger it. Includes job-search tracking as one topic area; read references/job-search.md when applications, deadlines, or interviews are the subject.

  • Installs packages

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read AGENTS.md and the memory index to locate the active vault.

    line 11
  2. Distinctive:

    If the task is about an ongoing job or internship search, also read references/job-search.md and keep its three layers apart.

    line 12
  3. Distinctive:

    Update only affected canonical files.

    line 14

≈ “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

Personal Secretary Memory

Support a long-running personal-secretary project through one factual memory source. The active vault and area are defined by the local project AGENTS.md; this public Skill does not assume a home-machine path.

  1. Read AGENTS.md and the memory index to locate the active vault. Read only the current status and task-relevant topic; do not load the whole vault. If the task is about an ongoing job or internship search, also read references/job-search.md and keep its three layers apart.
  2. For changes or exports, preserve event date, recording date, source type, uncertainty, latest corrections, and authorization boundaries.
  3. Update only affected canonical files. Historical packages and exported snapshots are evidence, not competing current-state stores.
  4. Separate user statements, original material, summaries, assistant suggestions, and reproducible tool checks. Never invent missing dates, metrics, credentials, or outcomes.
  5. For structural checks, run scripts/validate_memory.py (it needs PyYAML, install with pip install pyyaml first) with the actual vault and area. The check is read-only and does not prove historical truth or user acceptance.
  6. Report changed files, actual outputs, checks, and unresolved coverage. Do not claim automatic ingestion, cross-device sync, complete chat recovery, or acceptance without evidence.

Private memory stays local. Do not publish credentials, webhook values, tokens, identity documents, private chat transcripts, or unrelated company material.

Example

One entry, dated, with its source marked:

## 2026-10-07(记录日期 2026-10-07)
- 收到 A 公司内容运营岗笔试,已回信申请转评 AIGC 方向  [来源:本人陈述]
- 当前实习预计持续到 12 月中旬  [来源:offer 邮件]
- B 公司一面时间还没定  [来源:未确认,待核]

"未确认" is written as unconfirmed, never filled in with a plausible date.

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.