skills/context-use/SKILL.md
Personal memory from real data exports (ChatGPT, Claude, Instagram, Google, etc.). Search and manage memories extracted from the user's data before responding to anything personal — past conversations, people, places, projects, habits, or events.
npx skillsauth add onfabric/context-use context-useInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Install: pip install context-use
Requires OpenAI API key. Check: context-use config show. Set: context-use config set-key <key>.
Use this skill when the user:
Do not use when the question is purely technical, general knowledge, or already answered in this session.
If context-use memories list --limit 1 returns nothing, the user hasn't ingested data yet.
context-use pipeline --quick <provider> <path-to-zip>
context-use agent profile
Share the profile with the user — it builds trust and lets them correct anything wrong early.Always search memories before responding to anything personal. Weave results naturally — never announce "I found this in your memories."
context-use memories search "<query>" --top-k 5
context-use memories search "<query>" --from 2024-01-01 --to 2024-12-31
Use when: user references past conversations, projects, people, events; says "remember when…"; topic benefits from personal context. Skip when: purely technical question; general knowledge; already searched this session. If nothing relevant comes back, move on without mentioning it.
context-use memories list [--limit 20]
context-use memories get <uuid>
context-use memories create --content "..." --from 2024-03-01 --to 2024-03-31
context-use memories update <id> --content "..."
context-use memories archive <id1> <id2> --superseded-by <new-id>
Use when: user says "remember that I…"; memory needs correction; memories should merge. Skip when: info is transient or only relevant to current task.
context-use memories export [--format json] [--out path/to/file.md]
context-use agent synthesise # distill pattern memories from events
context-use agent profile # compile first-person user profile
context-use agent ask "..." # free-form memory task
Suggest synthesise after ingesting new data.
User must download their own export (ZIP, don't unzip):
| Provider | Export location | |-----------|----------------| | ChatGPT | Settings → Data Controls → Export Data | | Claude | Settings → Account → Export Data | | Instagram | Settings → Accounts Center → Download your information (JSON) | | Google | takeout.google.com → select products → ZIP |
context-use pipeline --quick <provider> <path-to-zip> # last 30 days
context-use pipeline <provider> <path-to-zip> # full history, batch API
Providers: chatgpt, claude, instagram, google.
Verify: context-use memories list --limit 5.
| Error | Action |
|-------|--------|
| "OpenAI API key not configured" | Ask user for key → context-use config set-key <key> |
| Rate limits (quick mode) | Drop --quick to use batch API |
| Export not ready | Normal for Instagram/Google — user waits |
| context-use not found | pip install context-use |
| No memories found | Guide user through Bootstrap |
| Unknown provider | Must be: chatgpt, claude, instagram, google |
| File not found | Verify path with ls, confirm ZIP location |
development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.
development
End-to-end Parallels smoke, upgrade, and rerun workflow for OpenClaw across macOS, Windows, and Linux guests. Use when Codex needs to run, rerun, debug, or interpret VM-based install, onboarding, gateway smoke tests, latest-release-to-main upgrade checks, fresh snapshot retests, or optional Discord roundtrip verification under Parallels.