Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add charlesjones-dev/claude-code-plugins-dev --skill workflow-claude-actionsgit clone --depth 1 https://github.com/charlesjones-dev/claude-code-plugins-devWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/charlesjones-dev/claude-code-plugins-dev/workflow-claude-actions)<a href="https://agentmods.dev/skills/charlesjones-dev/claude-code-plugins-dev/workflow-claude-actions"><img src="https://agentmods.dev/badge/skills/charlesjones-dev/claude-code-plugins-dev/workflow-claude-actions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/charlesjones-dev/claude-code-plugins-dev/workflow-claude-actions"><img src="https://agentmods.dev/badge/skills/charlesjones-dev/claude-code-plugins-dev/workflow-claude-actions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00106 | $0.03721 |
| Opus 5.5 | $0.00042 | $0.01488 |
| Sonnet 5.5 | $0.00021 | $0.00744 |
| Haiku 4.5 | $0.00011 | $0.00372 |
Grade A, and why
workflow-claude-actions scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 14d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/workflow-claude-actions - Claude Code GitHub Actions on the newest Opus
This builds on Claude Code's built-in /install-github-app. The result is two workflows:
- a review workflow that runs the
code-reviewplugin on every PR and posts its findings as inline comments; - a mention workflow that answers
@claudein issues, PR comments and reviews.
Both pass --model opus. The alias resolves to the newest Opus that the action's bundled Claude Code knows about, and anthropics/claude-code-action@v1 is a floating tag, so new Opus releases are picked up without editing the files. Pinning the action to a SHA or exact version, or setting ANTHROPIC_MODEL or ANTHROPIC_DEFAULT_OPUS_MODEL, stops that.
The bundled workflows are in ${CLAUDE_SKILL_DIR}/assets/: claude-code-review.yml and claude.yml.
This skill may create one branch, change workflow files under .github/workflows/, push the branch and open one PR, and it does that only after the user approves the diff in step 3. It never merges, never changes other files, and never changes repo settings or secrets. The user sets any secret themselves.
1. Look at the repo
-
gh repo view --json nameWithOwner,defaultBranchRef,viewerPermission,isPrivate. -
git status --porcelainmust be empty. If it isn't, stop and ask the user to commit or stash first. Record the current branch so you can switch back at the end. -
Base branch: if the repo's CLAUDE.md names an integration branch that work lands on, or recent PRs into the default branch come from a
staging(ordevelop) branch, target that branch. Otherwise target the default branch. -
Existing workflows: find the Claude workflows on the base branch by content, not filename:
git fetch origin <base> git grep -l "anthropics/claude-code-action" "origin/<base>" -- .github/workflows- A file that triggers on
pull_request(orpull_request_target) and reviews PRs is a review workflow. - A file that triggers on
issue_comment,pull_request_review_comment,pull_request_revieworissuesand checks for@claudeis a mention workflow. - One file can be both. It counts as both, so don't copy the other bundled file.
- Other Claude workflows (issue triage, CI fixes, scheduled jobs) are out of scope. Leave them alone, and mention them in the report.
- A file that triggers on
-
For each review or mention workflow, look at every step that uses the action (a file can have several jobs or steps) and note:
- The action ref.
@v1floats. A SHA or an exact version such as@v1.0.12is pinned. - Pre-v1.
@beta,@v0or anyv0.xref, or pre-v1 inputs (direct_prompt,override_prompt,model,allowed_tools,disallowed_tools,custom_instructions,mode,max_turns,claude_env,mcp_config,timeout_minutes) mean the old action, which has noclaude_argsinput. - The auth setup.
claude_code_oauth_token,anthropic_api_key, or another setup:use_bedrock,use_vertex,use_foundry, oranthropic_federation_rule_id(workload identity federation). The other setups also rely on env vars or credential steps. - The model.
claude_argsand any--model(or--model=) in it, and anyANTHROPIC_MODELorANTHROPIC_DEFAULT_OPUS_MODELin the step's, job's or workflow'senv:.ANTHROPIC_MODELoverrides--model. - For the review workflow, whether it posts to the PR. It does if it has
--commentin a/code-reviewprompt, themcp__github_inline_comment__create_inline_commenttool, or a custom prompt that comments withgh pr comment. - For the review workflow on a public repo, whether it skips fork PRs and bot PRs. Fork PRs get no OIDC token or secrets, and the action refuses PRs from bots such as Dependabot, so without a guard those PRs get a red check.
- Its state in
gh workflow list --all --json path,state.
- The action ref.
-
Claude GitHub App: a user token can't see whether the app is installed. A recent comment or review by
claude[bot]in the repo proves it is:
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 14d ago First seen · 140 lines · 106 tokens per session scan A e0b5bd1c8369
workflow-claude-actions is a skill published in the GitHub repository charlesjones-dev/claude-code-plugins-dev (35 stars, last pushed 14d ago), licensed MIT. It adds 106 tokens to every session and 3,721 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-11.
Other skills, from other repositories
green-prs
Inspect and prepare the selected PR queue against its exact current base, repair authorized defects with contributor credit, and report full versus partial validation. Use for maintainer pre-greening; merging requires separate owner authorization and the shared merge gates.
ai-qa-review
Review EXISTING test code for quality, smells, and testability issues. Detects test smells across six dimensions — readability, reliability, diagnostic value, design, AI-generated, and coverage — analyzes testability of application code, and backs the qualitative smells with mutation testing. Use when: "review my…
pr-babysitter
Monitors or repairs an open GitHub PR: CI failures, conflicts, review threads, and merge readiness, reporting state changes. Use when asked to "watch this PR", "fix CI", "resolve conflicts", or "address review comments".
tech-debt-ci-review
Codex adapter for deep technical-debt and CI-stability audits. Use when asked to find test theater, flaky tests, missing or mis-scoped tests, brittle CI/toolchain behavior, structural debt blocking green PRs, or a remediation order for opencode-swarm.
squid-implement-night
Run the full agent-team pipeline end-to-end for one feature whose Tasks Plan is already approved by /squid-plan, handing the human a validated, ready-to-squash-merge PR. Trigger after /squid-plan.
aster-review-ci
Run aster code reviews non-interactively in CI, GitHub Actions, or from another agent. Covers aster review --pr, --json, --stream, --comment, diff-from-stdin, token handling, and filtering findings. Use when wiring aster into a pipeline, posting PR comments, or parsing review output programmatically.