context-hygiene — Claude Code skill for context cleanup
This Claude Code skill archives old session log entries and trims active priorities. Run /context-hygiene to clean the context files.
MIT ★ 26 updated 2026-11-16
npx skills add wrg32786/aigent-os --skill context-hygiene -a claude-codeWhat context-hygiene does
The skill checks vault/memory/SESSION_LOG.md and vault/memory/ACTIVE_PRIORITIES.md, then runs node daemons/nightly-context-hygiene.mjs --root <aigent-root>. It hashes the source files before editing. If a hash changes, the run stops with CONTEXT_HYGIENE_RACE.
For the session log, it moves older complete blocks to an archive and keeps the five newest. In the priorities file, only content explicitly marked stale or superseded can be archived. The archive is written before the live file is trimmed, and the checker runs again afterward. Other memory files are left untouched. Node.js and the two memory files are required.
What it helps with
- archive older session log blocks
- keep the five newest dated entries live
- move stale priorities into an archive
- detect changes to source files during cleanup
Example prompts for the agent
- Run
/context-hygieneto clean up the session log. - Use
/context-hygieneto archive older session entries. - Run
/context-hygieneto trim active priorities.
How to install context-hygiene
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill context-hygiene -a claude-code2Manually: put the files in place
~/.claude/skills/context-hygiene/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/context-hygiene/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill context-hygiene -a codex2Manually: put the files in place
.agents/skills/context-hygiene/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill context-hygiene -a cursor2Manually: put the files in place
.cursor/skills/context-hygiene/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install wrg32786/aigent-os skills/context-hygiene2Manually: put the files in place
.github/skills/context-hygiene/SKILL.md1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill context-hygiene -a gemini-cli2Platform CLI
gemini skills install https://github.com/wrg32786/aigent-os --path skills/context-hygieneManually: put the files in place
.devin/skills/context-hygiene/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill context-hygiene -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about context-hygiene
How do I install context-hygiene in Claude Code?
Run npx skills add wrg32786/aigent-os --skill context-hygiene -a claude-code. The skill file is at ~/.claude/skills/context-hygiene/SKILL.md.
What does context-hygiene do?
It checks the session log and active priorities, moves eligible older entries into archive files, then runs the checker again.
What does context-hygiene need to run?
It requires Node.js and the files vault/memory/SESSION_LOG.md and vault/memory/ACTIVE_PRIORITIES.md.
Source text
SKILL.md · MIT
--- name: context-hygiene agent: none description: "Archive and trim ACTIVE_PRIORITIES and dated SESSION_LOG entries. Use when cleaning active priorities, trimming session logs, or running context hygiene." allowed-tools: Read, Write, Bash user-invocable: true status: PRODUCTION - bounded archive writer triggers: - context hygiene - trim session log - archive old session entries - clean active priorities --- # /context-hygiene This is the sole bounded compaction exception to the Session Log chronicle. It runs only when invoked, never while the framework is being installed or tested against a real vault. ## Precondition 1. Read `vault/memory/SESSION_LOG.md` and `vault/memory/ACTIVE_PRIORITIES.md`. 2. Hash both exact byte streams with SHA-256. 3. Run: ```text node daemons/nightly-context-hygiene.mjs --root <aigent-root> ``` Green means no edit is needed. Red permits only the bounded archive below. Unreadable or unfamiliar content is a failure, not permission to rewrite. ## Session Log archive The executable target schema uses newest-first dated H2 blocks. A canonical block starts with `## YYYY-MM-DD - title` and contains one nonempty occurrence of each field: - `Objective` - `Completed` - `Decisions` - `Open threads` - `Next action` Archival rules: - Keep the newest five live dated blocks. - Move older whole blocks to `vault/memory/SESSION_LOG_ARCHIVE_<YYYY-MM-DD>.md`. - Preserve source text and order. - Preserve frontmatter, title, schema guidance, and archive links. - Dedupe by exact dated heading plus block hash. - Refuse a block whose boundary cannot be determined without judgment. ## Active Priorities archive The mechanically checkable live shape contains: - exactly one `## Operating Mode: <mode>` heading; - one to five live bullet priorities under `## Tier 1`, `## Tier 2`, or `## Tier 3` headings; - no placeholder priority; - one valid `Last reviewed: YYYY-MM-DD` line; and - no embedded `## Archived` section. Move only content explicitly marked stale, superseded, or archived to `vault/memory/ACTIVE_PRIORITIES_ARCHIVE_<YYYY-MM-DD>.md`. If removal requires a human decision, preserve it and report the gate. ## Compare-and-swap Immediately before each write, hash the live source again. If either hash differs from the precondition hash, abort with `CONTEXT_HYGIENE_RACE`. Archive must be durable before the source is trimmed. After writing, re-run the checker. Only its literal `CONTEXT_HYGIENE PASS` receipt may record the checkpoint green. Return archive paths and before/after hashes in bounded detail. Never touch another memory file.
© wrg32786 · MIT · wrg32786/aigent-os