contribute
Local-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers a
npx skills add jeremylongshore/claude-code-plugins-plus-skills --skill contribute --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill provides a local command-center for OSS contributions. It maintains state in Markdown files, refreshes in-flight PR/issue context on invoke, helps find issues to contribute to on GitHub, and builds per-repo dossiers with gates to prevent sloppy AI contributions from reaching maintainers. It supports drafting claims, Design Issues, or PR descriptions, and runs through a gate-checked transition before any external action. It uses candidate files at ~/.contribute-system/candidates/ and dossiers at ~/.contribute-system/research/, with an append-only log at ~/.contribute-system/log.jsonl. It does not use a database or cloud calls and integrates with gh and jq on PATH.
How it works
- On activation, surface current state by querying GitHub for open/closed PRs, local candidate markdowns, and recent activity in the log. It relies on gh, jq, and shell commands to collect data from: candidates in ~/.contribute-system/candidates/.md, dossiers in ~/.contribute-system/research/.md, and log.jsonl. It instructs to run a parallel set of commands to list open PRs, closed PRs, and to summarize candidates and recent events.
- It maintains three state stores: (1) live GitHub state via gh, (2) candidate markdown files with frontmatter and body in ~/.contribute-system/candidates/, (3) repo dossiers in ~/.contribute-system/research/ as canonical gates (CLA, DCO, branch conventions, AI policy, draft-first, review bots, issue templates). It appends events to ~/.contribute-system/log.jsonl.
- Step 0 — Refresh state: run a series of gh and local file scans to enumerate in-flight PRs, recently merged/closed upstream PRs, local candidates, dossiers, profile, and log events. It excludes own-repo PRs by matching owner prefixes.
- Step 1 — Discover: delegate discovery to @scout to produce ranked candidate markdowns in ~/.contribute-system/candidates/ and log events; summarize top picks for the user.
- Step 2 — Qualify: run a set of gh API checks against a target repo (pr list, commits, CONTRIBUTING.md) and use repo-analyzer to assess eligibility and rules.
- Step 3 — Claim: draft a claim from assets/claim-template.md, show for user approval, gate-check via transition.sh, and only then update the candidate status. It does not auto-submit.
- Step 4 — Work and Step 5 — Submit: run local test suites per project stack and then draft and submit Design Issues or PRs with proper approvals. Submissions are gate-checked; steps are presented to the user for approval.
- Reconciliation step adjusts candidate status based on actual PR state from GitHub and updates frontmatter fields accordingly.
When to use it
Use when the user asks about their PRs/issues/contributions, wants to find new work, claim issues, refresh or build a dossier, draft a Design Issue or PR, or run an upstream repo's test suite. It is designed for contributions into repos the user does not own (anti-slop OSS maintenance). If the user references own-org repos, that is treated as a scout bug and should be flagged.
What it can touch
- Candidate files: ~/.contribute-system/candidates/*.md
- Repo dossiers: ~/.contribute-system/research/*.md
- Event log: ~/.contribute-system/log.jsonl
- Dossier creation/refresh delegated to @researcher and gate-checker via transition.sh
- Discovery: uses ~/.contribute-system/profile.md and external gh search results
- Commands and scripts: references to gh, jq, transition.sh, assets templates, and various Bash pipelines as described in Step instructions.
Caveats
- Requires gh CLI and jq on PATH, plus a workspace at ~/000-projects/contributing-clanker/ with upstream clones.
- State is markdown-based with no database or cloud backend; data is kept in local files and live GitHub state.
- Designed to operate on contributions into repos not owned by the user; own-repo contributions are out of scope and filtered.
- Human approval is mandatory before external submissions, per the repository policy cited in the skill.
# Contribute Command Center ## Overview Local-only OSS contribution workflow. The skill itself is the system — there is no separate CLI binary, dashboard, or cloud backend. State lives in three places: 1. **GitHub itself** — fetched live via `gh` for any PR/issue state question. Never cached long-term. 2. **Markdown candidate files** at `~/.contribute-system/candidates/<owner>__<repo>__issue<N>.md
What does the contribute skill do?
Local-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers a
How do I install it?
Run `npx skills add jeremylongshore/claude-code-plugins-plus-skills --skill contribute --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From jeremylongshore/claude-code-plugins-plus-skills, a repository with 2,596 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.