Workflows

The full workflow gallery.

GitHub Actions you can paste into a repo. Filter by the coding agent that runs and by what starts the run, then open a workflow for the YAML and the setup.

workflows: 63; workflows
agents: 5; agents
triggers: 8; triggers
use skills: 26; use skills

Agent

All63 Claude24 Codex15 Cursor2 gh-aw21 Multi-agent1

Trigger

All63 Assignment1 CI failure6 Comment9 Issue opened4 Label9 Other5 Pull request10 Schedule19

63 workflows

ClaudePull request
01
Claude auto PR review
Copied from the official Solutions Guide basic example (prompt shortened).
pull_request: opened, synchronize
Est. cost per run~$0.18Checked Oct 16, 2026
Not executed
ClaudePull request
02
Claude PR review with progress tracking
`track_progress: true` posts a checkbox tracking comment (per Solutions Guide).
pull_request: opened, synchronize, ready_for_review, reopened
Est. cost per run~$0.22Checked Oct 16, 2026
Not executed
ClaudeComment
03
@claude mention assistant on issues and PRs
No `prompt` → the action runs in tag/interactive mode and responds to the comment. `trigger_phrase` default is `@claude` (usage.md).
issue_comment / pull_request_review_comment containing @claude
Est. cost per run~$0.27Checked Oct 16, 2026
Not executed· 1 caveat
ClaudeLabel
04
Implement an issue when a label is applied
`label_trigger`: 'The label name that triggers the action when applied to an issue' (usage.md).
issues: labeled (label_trigger)
Est. cost per run~$0.69Checked Oct 16, 2026
Not executed· 1 caveat
ClaudeAssignment
05
Implement when an issue is assigned to a bot user
`assignee_trigger`: username that triggers on issue assignment.
issues: assigned (assignee_trigger)
Est. cost per run~$0.69Checked Oct 16, 2026
Not executed· 1 caveat
ClaudeIssue opened
06
Auto-triage and label new issues
Requires `scripts/gh.sh` and `scripts/edit-issue-labels.sh` copied from the action repo's scripts/ dir (per Solutions Guide).
issues: opened
Est. cost per run~$0.06Checked Oct 16, 2026
Not executed· 1 caveat
ClaudePull request
07
Review only when critical paths change
Pattern from Solutions Guide 'Review Only Specific File Paths'; paths are examples.
pull_request with paths filter
Est. cost per run~$0.18Checked Oct 16, 2026
Not executed· 1 caveat
ClaudePull request
08
OWASP-focused security PR review
`use_sticky_comment` keeps one updatable comment (usage.md).
pull_request: opened, synchronize
Est. cost per run~$0.22Checked Oct 16, 2026
Not executed
ClaudeSchedule
09
Weekly repo health check
Based on Solutions Guide 'Scheduled Repository Maintenance'.
schedule (cron) + workflow_dispatch
Est. cost per run~$0.44Checked Oct 16, 2026
Not executed· 1 caveat
ClaudePull request
10
Update docs when API code changes
Based on Solutions Guide 'Documentation Sync on API Changes'.
pull_request with paths: src/api/**
Est. cost per run~$0.35Checked Oct 16, 2026
Not executed· 1 caveat
ClaudeCI failure
11
Read failing CI logs and suggest a fix
`additional_permissions` 'Currently supports actions: read' (usage.md).
workflow_run: completed (failure)
Est. cost per run~$0.35Checked Oct 16, 2026
Not executed· 1 caveat
CodexPull request
12
Codex PR review posted as a comment
Two-job split keeps the API key job read-only (README example pattern). `:read-only` profile per README Permission profiles.
pull_request: opened
Est. cost per run~$0.03 runner + OpenAI tokens (UNVERIFIED)Checked Oct 16, 2026
Not executed· 3 caveats
CodexLabel
13
Codex implements an issue on label
`:workspace` profile lets Codex edit the checkout (README). PR creation is plain git/gh, not an action input.
issues: labeled == codex
Est. cost per run~$0.09 runner + OpenAI tokens (UNVERIFIED)Checked Oct 16, 2026
Not executed· 2 caveats
CodexIssue opened
14
Codex triage with JSON output schema
`output-schema` is forwarded to `codex exec --output-schema` (README inputs table).
issues: opened
Est. cost per run~$0.01 runner + OpenAI tokens (UNVERIFIED)Checked Oct 16, 2026
Not executed· 3 caveats
CursorLabel
15
Launch a Cursor Cloud Agent when an issue is labeled
Fields `prompt.text`, `repos[].url`, `repos[].startingRef`, `autoCreatePR` from the v1 Create An Agent docs. API is public beta.
issues: labeled == cursor
Est. cost per run~$0.01 runner + Cursor plan usage (UNVERIFIED)Checked Oct 16, 2026
Not executed· 2 caveats
CursorLabel
16
Cursor agent drafts a plan first (plan mode)
`mode: plan` 'explores and drafts a plan before coding' (v1 docs).
issues: labeled == planning
Est. cost per run~$0.01 runner + Cursor plan usage (UNVERIFIED)Checked Oct 16, 2026
Not executed· 2 caveats
gh-awIssue opened
17
gh-aw: agentic issue triage with safe label output
gh-aw source is Markdown + YAML frontmatter compiled with `gh aw compile` to a .lock.yml. Writes go through `safe-outputs`.
issues: opened
Est. cost per run~$0.02 runner + engine tokens/credits (UNVERIFIED)Checked Oct 16, 2026
Not executed· 2 caveats
gh-awLabel
18
gh-aw: run an AI review when `ai-review` label is added
`label_command` removes the label after activation by default (`remove_label: true`).
label_command: ai-review
Est. cost per run~$0.03 runner + engine tokens/credits (UNVERIFIED)Checked Oct 16, 2026
Not executed· 1 caveat
gh-awSchedule
19
gh-aw: daily CI optimization coach (GitHub's own)
Trimmed from gh-aw's own .github/workflows/ci-coach.md (removed imports, experiments, AI-credit caps).
schedule: daily + workflow_dispatch
Est. cost per run~$0.12 runner + engine tokens/credits (UNVERIFIED)Checked Oct 16, 2026
Not executed· 1 caveat
Multi-agentLabel
20
Label-driven dev loop: triage → plan → implement → red-team → human merge
Mirrors Greg's Trailside Dev Loop labels. Different vendors for author vs red-team reduce shared blind spots. Human gates: adding `ready` approves the plan; merge is manual.
issues/pull_request: labeled
Est. cost per run~$0.40–$2.50 per issue end to end (ASSUMPTION)Checked Oct 16, 2026
Not executed· 3 caveats
gh-awSchedule
21
Accessibility review of web UI changes
Runs daily. The agent checks your web app against WCAG 2.2 accessibility guidelines and reports the problems it finds.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
gh-awCI failure
22
CI Doctor: investigate failed CI runs and open a diagnosis issue
When CI fails on main, the agent reads the job logs and repository context, finds the likely root cause, and opens one diagnosis issue with suggested fixes, merging duplicate failures together.
workflow_run: CI completed on main (failure)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
23
Code simplifier: PRs that simplify recently changed code
Runs daily. The agent looks at recently changed code and opens pull requests that simplify it without changing behavior.
schedule: daily
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
24
Contribution check: review incoming PRs against contributor rules
Every four hours, the agent reviews incoming pull requests against your contribution guidelines, labels them (spam, needs-work, outdated or lgtm), comments, and files a summary report issue.
schedule: every 4 hours + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
25
Daily doc updater: keep docs in sync with merged code
Runs daily. The agent compares recently merged code with the documentation and opens a pull request that updates docs that have drifted.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
26
Daily malicious code scan of recent changes
Runs daily. The agent reviews the last three days of code changes for patterns that suggest malicious code or agent-driven attacks, and reports anything suspicious.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
27
Daily perf improver: agent finds and fixes performance issues
Runs daily (or on /perf-assist). The agent looks for performance problems, measures them, and opens pull requests with fixes, with a cap on open PRs.
schedule: daily + workflow_dispatch + /perf-assist command
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
28
Daily repo status report issue
Runs daily. The agent writes a status report on recent repository activity (issues, pull requests and progress) and posts it as an issue.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
gh-awSchedule
29
Daily test improver: agent adds tests and raises coverage
Runs daily (or on /test-assist). The agent finds gaps in test coverage, writes new tests, and opens pull requests that raise coverage, with a cap on open PRs.
schedule: daily + workflow_dispatch + /test-assist command
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
30
Duplicate code detector: flag copy-paste and suggest refactors
Runs daily. The agent scans the codebase for copy-pasted or near-duplicate code and opens an issue suggesting refactors.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awComment
31
Grumpy reviewer: critical edge-case code review on command
Comment /grumpy on a pull request to get a deliberately critical review that hunts for edge cases, likely bugs and code-quality problems.
slash_command: /grumpy comment on a PR
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
32
Issue arborist: link related issues as sub-issues
Runs daily. The agent reads open issues and links related ones together as sub-issues so the backlog is easier to navigate.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
33
Link checker: find and fix broken doc links daily
Runs on weekdays. The agent finds broken links in your documentation and opens a pull request that fixes them.
schedule: daily on weekdays
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
34
Project planner: daily roadmap discussion from issues and PRs
Runs daily. The agent reviews open issues, pull requests and recent work, then creates or updates a planning discussion with a prioritized roadmap and suggested new issues. It reads maintainer comments on the plan.
schedule: daily + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
gh-awComment
35
PR Fix: /pr-fix command makes the agent fix a failing PR
Comment /pr-fix on a pull request and the agent looks at the failing checks, finds the root cause, and pushes a fix to the PR branch.
slash_command: /pr-fix comment
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
gh-awComment
36
Nitpick reviewer: /nit for detailed style and best-practice review
Comment /nit on a pull request to get a detailed review of style, best practices and small improvements.
slash_command: /nit comment on a PR
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awComment
37
Q: workflow optimizer agent for your agentic workflows
Comment /q to ask an expert agent about your agentic workflows. It answers questions, analyzes the repo, and can open pull requests that optimize your gh-aw workflows.
slash_command: /q comment
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
gh-awSchedule
38
Repo Assist: a daily repository helper agent
A repository helper that runs twice a day and on /repo-assist. It triages issues, answers questions, and opens small pull requests for maintainers, with a cap on open PRs.
schedule: every 12h + workflow_dispatch + /repo-assist command
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 2 caveats
CodexPull request
39
Block: Codex security review on PRs with an isolated read-only job
Codex security-reviews pull requests to main in an isolated read-only job, then posts an advisory review. Untrusted PRs are only reviewed after a maintainer comment command. Used in Block's buzz repository.
pull_request_target: opened, reopened, ready_for_review, synchronize (+ push to main, repository_dispatch, issue replies)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudePull request
40
Agent approval check: require human approval on PRs with agent-authored commits
Sets a required status check that fails until a human approves any pull request containing agent-authored commits. It runs from the default branch, so a PR cannot edit the check to approve itself. Approvals are re-read on new pushes or an /approve comment.
pull_request_target: opened, synchronize, reopened, ready_for_review (re-checked on new PR replies)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeCI failure
41
CI failure auto-fix: Claude pushes a fix branch when CI fails
When CI fails on a pull request, Claude reads the failure logs, makes a fix, and pushes it to a new branch for you to review. The upstream file warns that it runs PR code with write permissions, so use it only on trusted branches.
workflow_run: CI completed (failure)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeComment
42
@claude assistant via Workload Identity Federation (no long-lived API key)
The standard @claude assistant for issues, PR comments and reviews, but it signs in with GitHub OIDC (Workload Identity Federation) instead of a long-lived ANTHROPIC_API_KEY secret. It needs a one-time federation setup in the Claude Console.
issue_comment / pull_request_review_comment containing @claude (+ issues: opened, assigned; reviews)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeIssue opened
43
Issue deduplication: Claude finds and flags duplicate issues
When a new issue is opened, Claude searches existing issues for duplicates. If it finds one, it comments with a link and applies a duplicate label.
issues: opened
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeOther
44
Claude commit analysis on demand (workflow_dispatch)
A manual workflow you start from the Actions tab. Claude either summarizes the latest commit or security-reviews it, depending on the option you pick.
workflow_dispatch (manual: summarize-commit or security-review)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudePull request
45
Comprehensive PR review with inline comments
Claude reviews each pull request and leaves inline comments on specific lines, with a progress-tracking comment that updates while the review runs.
pull_request: opened, synchronize, ready_for_review, reopened
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudePull request
46
PR review only for selected authors (e.g. external contributors)
Runs a Claude PR review only for pull requests from authors you list, for example external contributors. Everyone else's PRs are skipped.
pull_request: opened, synchronize (selected authors only)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeCI failure
47
Auto-retry flaky tests: Claude classifies CI failures and retries flaky ones
When CI fails, Claude uses structured output to decide whether the failure looks flaky. Flaky runs are retried automatically; real failures are left for a human.
workflow_run: CI completed (failure)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexOther
48
Test Codex sandbox protections against secret exfiltration
A manual test showing why Codex's drop-sudo and read-only safety strategies matter. It checks that the agent cannot read or exfiltrate API keys from the runner.
workflow_dispatch (manual)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexOther
49
Run Codex as an unprivileged user (hardened runner setup)
Example setup that runs the Codex action as an unprivileged user on the runner, so the agent cannot use sudo or read the runner's secrets. You start it manually.
workflow_dispatch (manual)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeComment
50
Firezone: @claude assistant on issues, PR comments and reviews
Mention @claude in an issue, a PR comment or a review and Claude responds or makes the requested change. This is Firezone's production setup.
issue_comment / pull_request_review_comment containing @claude (+ issues: opened; reviews)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeLabel
51
Foam: Claude drafts a spec from an issue
Label an issue agent:spec and Claude drafts a spec file for it and opens it as a draft pull request for review.
issues: labeled == agent:spec (+ workflow_dispatch)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexSchedule
52
Guardian: Codex-driven incremental TypeScript migration
Runs daily. Codex migrates a small batch of JavaScript files to TypeScript and opens a pull request, so a large migration happens incrementally.
schedule: daily (cron) + push + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexSchedule
53
Jitsu: weekly Codex run that fixes security vulnerabilities and opens PRs
Runs weekly. Codex finds dependency security vulnerabilities, fixes those above a severity threshold, and opens pull requests. It supports a dry-run mode that only reports.
schedule: weekly (cron) + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexSchedule
54
openai-node: monthly Codex review of supported Node.js versions
Once a month, Codex compares the repository's supported Node.js versions with the official Node.js release schedule. If anything has drifted, it makes the edits, validates them, and opens a draft PR for human review.
schedule: monthly (cron) + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexSchedule
55
OpenClaw: sweep expired dated TODOs in the codebase
Runs weekly. Codex finds TODO comments whose dates have passed and keeps a tracking issue listing them. It supports a dry run. Used in the OpenClaw repository.
schedule: weekly (cron) + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexOther
56
OpenClaw: docs agent repairs docs after CI passes on main
After CI passes on main, Codex checks whether the docs still match the code and repairs them, at most once an hour. Used in the OpenClaw repository.
after CI succeeds on main (workflow run event) + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexCI failure
57
Codex autofix when CI fails
When CI fails, Codex reads the failure, makes a fix, and opens a pull request with it.
workflow_run: CI completed (failure)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeComment
58
Pulumi registry: @claude assistant on issues, PRs and reviews
Mention @claude in an issue, a PR comment or a review and Claude responds or makes the requested change. This is the Pulumi registry's production setup.
issue_comment / pull_request_review_comment containing @claude (+ issues: opened, assigned; reviews)
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexComment
59
uv: Codex refreshes an issue's context summary on new comments
When someone comments on an issue, Codex folds the useful parts into the issue's maintainer context summary so it stays current. Used in the uv repository.
issue_comment: created
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeLabel
60
vcad: label-driven Claude agent that implements issues
Label an issue agent:ready and Claude implements it and opens a pull request. The labels move to agent:running while it works.
issues: labeled == agent:ready
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
ClaudeCI failure
61
vcad: Claude fixes failing CI automatically
When CI fails on a pull request that has the auto-fix label, Claude pushes a fix commit to the branch. The number of auto-fix attempts per branch is capped.
workflow_run: CI completed (failure), opted-in PRs only
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexOther
62
video.js: Codex writes changelog prose when a release is published
When a video.js core release is published, Codex writes readable changelog prose for it. It can also be run manually to backfill an older release.
release: published + workflow_dispatch
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat
CodexLabel
63
video.js: Codex turns a labeled issue into a PR
Label an issue agent:pr and Codex implements it and opens a pull request. Used in the video.js repository.
issues: labeled == agent:pr
Est. cost per runUnknown (not estimated)Checked Oct 17, 2026
Not executed· 1 caveat