Agent skill · software engineering · steipete
github-cache-hygiene
GitHub quota/cache hygiene: Gitcrawl archives, Octopool-backed gh, freshness, limits.
Why this skill is useful
Adds specific commands and patterns for optimizing GitHub API usage that the AI wouldn't reliably generate on its own.
What it needs
Requires gh, gitcrawl, octopool installed locally. About 4k tokens when loaded. Last updated 2026-08-06. 6,510 stars on the source repository.
What this skill does
GitHub Cache Hygiene Goal: discover in the local Gitcrawl archive first, then use the existing Octopool-backed gh shim for current GitHub metadata and authorized writes. Default Path Start with local archive reads: --include-closed keeps closed or merged candidates in scope. Archive state can lag GitHub; it is not proof of current state. Then use bare PATH gh when current metadata is needed. On Peter's machines it is expected to be the Octopool-backed shim, so supported JSON reads share the fleet cache without changing authentication or command routing: Use exact refs and narrow fields. --json selects output, not a reader identity. In Octopool 0.7.1, PR mergeStateStatus and issue comments select native gh for the entire field bundle. PR comment/review/commit exports also retain native handling. Request those fields when needed, but omit them from descriptive reads that do not consume them. Never remove merge/review gates merely to raise cache reuse, or split one decision into separately timed reads without preserving its consistency contract. Supported human output and recognized landing GraphQL projections can also relay. Supported gh api --paginate/--slurp and --hostname github.com reads need not bypass Octopool. Use the installed version's CLI contract and observed routing; an absolute gh path alone does not establish that it is the native binary. Avoid broad loops like one gh issue view per result when a single gh search or gh issue list --json ... can answer the first-pass question. For CI, avoid tight gh run list / gh run view polling loops. After a push or workflow dispatch, identify one exact run, then use an exact-run watcher or poll that run at 30s, 60s, then 120s intervals. Check the expected head SHA and run attempt. A broad or filtered list can lag even on a fresh upstream acquisition; its newest row is not final CI evidence. Fetch logs once, only after failure or explicit request. Reuse prior output instead of re-reading completed runs. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills steipete/github-cache-hygiene