Local skills help AI agents keep long work on track
COMPASS Skills is a set of four local skills for that need to continue long tasks across conversations. The four skills help clarify unclear work before action, keep a repo-local record of task progress, create a for a new AI chat, and store a local collaboration profile for . Task records usually live inside the project at `.agent-workbench/task-forest/`, and user profiles live under `.compass-skills/user-profiles/v1` by default.
The skills can be installed for Claude Code and Codex with commands such as `npx skills add dongshuyan/compass-skills --skill '*' -a codex -a claude-code`. The system is designed to keep runtime data local and says it does not extract , tokens, private keys, credentials, or sessions. The user profile is stored as plain text without encryption, so passwords, tokens, codes, and highly sensitive personal data should not be saved there.
The public install path has been checked with `skills@1.5.11`, including listing skills, installing them into Claude Code, and running a handoff smoke test.
Key points
- The project ships four skills: task clarification, task memory, s, and user profile management.
- It records long-running work inside the repository so progress can survive chat boundaries.
- It can create a shorter instead of replaying a full .
- It supports Claude Code and Codex through the skills installer or manual folder copying.
- Local profile data is plain text, so secrets and sensitive personal details should stay out.