# Review Playbook

> Review a commit or MR against the repo's playbook rules. Spawns a fresh playbook-reviewer agent that reads the rules and diff independently. Strict — every violation cited with file and line, pre-existing violations included.

- **Type:** Skill
- **Install:** `agentstack add skill-the-alien-club-claude-workflows-review-playbook`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [the-alien-club](https://agentstack.voostack.com/s/the-alien-club)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [the-alien-club](https://github.com/the-alien-club)
- **Source:** https://github.com/the-alien-club/claude-workflows/tree/main/skills/review_playbook

## Install

```sh
agentstack add skill-the-alien-club-claude-workflows-review-playbook
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Review Against Playbook

This skill spawns the `playbook-reviewer` agent with a target and a playbook path. The agent does all the reading — rules and diff — on its own. Your job here is only to resolve the target and locate the correct playbook.

---

## Step 1 — Resolve the Target

Argument forms:
- No argument → current branch vs `dev` (fall back to `main`)
- `!1234` or `1234` (numeric) → GitLab MR number
- 7–40 hex chars → commit SHA
- A branch name → that branch vs `dev`

---

## Step 2 — Find the Playbook

```bash
find . -type d -name playbook | grep -v node_modules | grep -v .git
```

If only one playbook exists, use it.

If multiple, determine which repo the target touches:
- Commit: `git show [sha] --name-only`
- Branch/MR: `git diff dev...[branch] --name-only`

Match changed paths to the correct repo subdirectory and pick its playbook.

If no playbook found: stop and report:
> No playbook found. Run `/add_playbook_rule` to create rules first.

---

## Step 3 — Spawn the Reviewer

Call `Agent` with `subagent_type: "playbook-reviewer"` and a prompt that contains only:
- The target identifier (commit SHA, branch range, or MR number)
- The absolute path to the playbook directory
- The working directory (so git commands resolve correctly)

Example prompt:

```
Review target: git diff dev...feature/my-branch
Playbook: /home/xqua/Documents/Work/Alien/DataStreaming/datastreaming-demos/lds-chatbot/playbook
Working directory: /home/xqua/Documents/Work/Alien/DataStreaming/datastreaming-demos/lds-chatbot
```

---

## Step 4 — Relay Output

Print the agent's output verbatim. Do not editorialize, soften, or add context.

Append this footer:

```
---
Playbook : [path]
Target   : [resolved target]
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [the-alien-club](https://github.com/the-alien-club)
- **Source:** [the-alien-club/claude-workflows](https://github.com/the-alien-club/claude-workflows)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-the-alien-club-claude-workflows-review-playbook
- Seller: https://agentstack.voostack.com/s/the-alien-club
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.