# Engineering templates

Code review, release notes, dependencies and repository chores.

- [Code review](https://marketplace.agentrun.ai/templates/code-review.md): Reviews a pull request on its own read-only checkout with eight independent reviewers (five quick, three deep), each with a fresh context and the shell where there is one, for correctness against the repo's CODEREVIEW.md. Each tries to break each changed function and quotes the line behind each finding; the agent merges and dedupes their findings, verifies them, ranks them P0/P1/P2 with a 1-5 risk score, posts one COMMENT review and answers with the findings as JSON.
- [Release notes](https://marketplace.agentrun.ai/templates/release-notes.md): Turns the pull requests merged since the last release into user-facing release notes (what you can do now, fixed, upgrading, changes by pull request), leaves them as a draft issue for a person to edit, and never publishes.
- [Dependency health check](https://marketplace.agentrun.ai/templates/dependency-health.md): Should you adopt this package, or keep the one you have? Give it a package name or a repository and get a dated health check, with the numbers from the registry, the repository and the advisory databases. It covers releases and maintenance, how many people really maintain it, its open and past CVEs with the fixed versions, its license and any relicensing, downloads and dependents, and two to four alternatives. You get a verdict (adopt, adopt with caution, keep but plan an exit, avoid), and every number is cited. Needs no GitHub connection, since it reads public GitHub anonymously.
- [Issue triage](https://marketplace.agentrun.ai/templates/issue-triage.md): Triages each new issue on your GitHub repository the moment it is opened. It labels the issue from the labels your repo already has, checks your open and closed issues for likely duplicates and links them, and asks the author in one comment for whatever a maintainer would need first (steps to reproduce, the version, the error). It can't close, edit or mark an issue a duplicate, and never creates a label. Connects through Composio's GitHub toolkit. Each label and comment waits for your OK unless you pick Never for approvals at install; it stays held to your repository either way.
- [Linear sprint check](https://marketplace.agentrun.ai/templates/linear-sprint-check.md): Twice a week, a Slack post on how your Linear team's current cycle is going. It lists the issues that are slipping (past their due date, stuck in progress, not started with half the cycle gone), the work added after the cycle started, and the issues whose linked GitHub pull request is waiting on a review or has gone quiet. Every line names the issue, who has it and why it's flagged, and things already flagged at the last check are collected on one line instead of repeated. It only reads Linear and GitHub. It posts on its own only if you pick Never for approvals at install; it stays held to your team and channel either way.
- [Linear triage](https://marketplace.agentrun.ai/templates/linear-triage.md): Triages each new issue in one Linear team as it arrives. It labels the issue from the team's own labels, checks the team's open and recent issues for likely duplicates, and posts one comment that links them and asks for whatever is missing (steps to reproduce, the version or browser, the error). It never closes, assigns, moves, prioritizes or creates anything. Connects through Composio's Linear toolkit. Each label and comment waits for your OK unless you pick Never for approvals at install; it stays held to your team either way.
- [Open-source project tracker](https://marketplace.agentrun.ai/templates/oss-tracker.md): A weekly health snapshot of the open-source projects you depend on or compete with. For each repository or package you get stars and their change, downloads and their trend, active contributors over 30 and 90 days, how concentrated the commits are (bus factor), issue and pull request health, the last release and the release pace, and anything that changed the picture this week (a major release, a CVE, a relicense, an archive, a call for maintainers). Each project is marked rising, stalling or fading on several signals, never one, and every number names its source and period. Needs no GitHub connection, since it reads public GitHub anonymously.
- [Release shipped](https://marketplace.agentrun.ai/templates/release-shipped.md): When you publish a release on GitHub, posts one short celebration to your team's Slack channel. It says what shipped in plain words, taken from the release notes and the pull requests merged for it, links the release and thanks the people who wrote it, by name and without pinging anyone. Pre-releases stay quiet unless you want them, and a release is never announced twice, even when you edit it. It watches the repository's release feed every 15 minutes, so it works on public repositories only (a private repository's feed needs a login). It reads GitHub with one fine-grained token, Contents read-only on that repository, added as the GitHub (token) server with Read-only on. It posts on its own only if you pick Never for approvals at install; it stays held to your repository and channel either way.
- [Stale PR digest](https://marketplace.agentrun.ai/templates/stale-pr-digest.md): Once a week, a Slack post listing the open pull requests in one GitHub repository that have sat with no review or no activity for longer than you allow, oldest wait first. Each line says who is waiting on whom (a requested reviewer who hasn't looked, an author who hasn't answered a change request, an approved PR nobody merged), how long, and a link. Drafts stay out. A PR still stuck from last week is marked as such instead of reported fresh. It only reads GitHub, through Composio's GitHub toolkit, and posts to one channel. It posts on its own only if you pick Never for approvals at install; it stays held to your repository and channel either way.
- [Vendor outage watch](https://marketplace.agentrun.ai/templates/vendor-outage-watch.md): A weekly brief on the outages, incidents and postmortems at the vendors your systems depend on (cloud, CDN, auth, payments, developer tools), read from each vendor's own status page and incident history. Each incident gives what broke, where, when it started and ended per the status page, what the vendor confirmed, the root cause once published, and what it means for your on-call. Planned maintenance and five-minute blips stay out, nothing repeats from last week, and a quiet week says so in one line. Optionally, give it your vendors' status feeds and it also alerts on each incident as the status page posts it, and again when its status changes; leave the list empty and it is the weekly brief alone.
- [Platform change log](https://marketplace.agentrun.ai/templates/platform-changes.md): A weekly list of the breaking changes, deprecations, shutdown dates, price and quota changes, policy and license changes in the platforms and APIs you build on (AWS, Stripe, OpenAI, Vercel and the like), read from each vendor's own changelog and deprecation pages. Each change gives the exact version, plan or region, the deadline as the vendor words it, and what to do about it in your stack. Changes that can't touch your setup and marketing posts stay out, and nothing repeats from last week.
- [PR alert](https://marketplace.agentrun.ai/templates/pr-alert.md): Tells you when someone opens a pull request on a public GitHub repository, one alert per new PR, saying who sent it (a first-time contributor, a regular or a member), what it is about, which files and areas it touches, and what to watch, such as CI, auth, dependency or release changes. It checks the repository's open pull requests every 30 minutes. It is not a code review, and it runs for $0 on a free OpenRouter model with no GitHub token, so it reads public repositories only.
- [Upgrade watch](https://marketplace.agentrun.ai/templates/upgrade-watch.md): Every Monday, reads what a project you depend on released since the version you run, checks each breaking or behavior change against the list of what your project uses from it, and answers safe to upgrade or names what breaks and what to do, each finding quoted from the release notes and linked. Notes it could not read are said, never guessed.
