Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -123,6 +123,16 @@ catalog/
`.lock.yml` files are marked `linguist-generated` and `merge=ours` in `.gitattributes` to prevent spurious merge conflicts.
</details>

## Publishing

> Maintainers only — requires write access to this repo.

1. **Bump the version** in `.claude-plugin/plugin.json`. Open a PR, get it merged to `main`.
2. **Create a GitHub release** tagged `vX.Y.Z` matching the new version. The self-hosted marketplace URL (`https://raw.githubusercontent.com/verkyyi/github-agent-runner/main/.claude-plugin/marketplace.json`) always resolves to the latest `main`, so there is no separate marketplace submission step — users who have already added the marketplace get the updated plugin on their next `/plugin update`.
3. **Update external listings** if the plugin description or category changed:
- [claude-plugins.dev](https://claude-plugins.dev) — submit the updated listing via the site's form.
- [ClaudePluginHub](https://claudepluginhub.com) — same.

## Credits

Built on two open-source projects from the [GitHub Next](https://githubnext.com) team:
Expand Down
2 changes: 1 addition & 1 deletion catalog/agent-team/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -111,7 +111,7 @@ Then apply the OAuth token tweak to each `.lock.yml` per [`skills/install-workfl

- **Concurrency**: each workflow uses `concurrency: group: agent-team-issue-${issue_number}` so only one role runs at a time per issue.
- **Max iterations**: default 3 (reviewer kickback → implementer). The counter lives on the `iteration` input passed through the dispatch chain, bumped exclusively by the reviewer on kickback.
- **Input propagation**: planner / implementer / reviewer must fail loudly if required `workflow_dispatch` inputs are missing. Do not rely on label search or recent-activity inference as a fallback.
- **Input propagation**: planner / implementer / reviewer fail loudly if required `workflow_dispatch` inputs are missing, whitespace-only, or arrive as unresolved template literals (e.g. `${{ github.event.inputs.issue_number }}`). When the guard triggers: if `issue_number` is available, the agent adds `state:blocked` to the issue and posts `🛑 agent-team: workflow_dispatch inputs were not propagated. Re-dispatch with valid inputs.`; if `issue_number` is itself absent, the agent calls `missing_data` / `report_incomplete` and stops. **Recovery**: open the upstream dispatch step in the Actions tab to find the actual values, then re-dispatch manually — `gh workflow run <next-agent>.lock.yml -f issue_number=<N> -f iteration=<N>`. Never infer missing inputs from labels or recent activity; that masks the root cause.
- **Non-UI only**: no screenshot capture. Reviewer validates via tests/CI status + reading the diff.
- **Cost**: a single task can easily spend 4× the tokens of a monolithic workflow. Set `timeout-minutes` conservatively and monitor the first few runs.
- **No auto-merge**: the reviewer approves but never merges. Humans merge.
Expand Down