agentsclimarketplace

Review playbook

Skill 3A2DEV/ansible-designer/skills/review-playbook

Claude code ansible skill

Install
npx -y skills add 3A2DEV/ansible-designer --skill review-playbook

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Review an existing Ansible playbook and produce a structured severity report grouped by CRITICAL, WARNING, and INFO. Triggered by /review-playbook. Checks FQCN usage, idempotency patterns, no_log on secret tasks, tag coverage, deprecated syntax, become usage, and style consistency. NEVER modifies files.

SKILL.md

5.9 KB, as published. Nobody here has run it

review-playbook

Review an Ansible playbook and produce a structured severity report. This command never modifies files.


Required Inputs

  1. path + filename — Path to the playbook to review (resolved from discovery if not provided)

Behavior

Step 1 — Discovery

Run discovery per references/discovery.md. Use context to:

  • Locate the playbook if only a filename is given (search in ./, ./playbooks/)
  • Identify available roles/collections for FQCN verification

Step 2 — Load and Analyze

Read the playbook file. Analyze every task, handler reference, role include, and play-level attribute.

Step 3 — Generate Severity Report

Produce a structured report in this exact format:

## Playbook Review: <filename>
Reviewed: <timestamp>
Plays: <count>  Tasks: <count>  Roles: <count>

---

### CRITICAL
[List of critical issues — must fix before use in production]

### WARNING
[List of warnings — should fix; may cause problems under certain conditions]

### INFO
[Informational suggestions — best practices, performance hints]

---
Summary: <X> critical, <Y> warnings, <Z> info

If there are no issues in a severity level, show: No issues found.


Checks to Perform

CRITICAL checks

CheckWhat to look forMessage format
Bare module namesAny task using a module without FQCN prefix[tasks:<line>] Task "<name>" uses bare module name '<module>' — must use FQCN '<ansible.builtin.module>'
Missing no_log on secretsTasks with password:, secret:, token:, key: arguments, or tasks referencing vault_* variables, without no_log: true[tasks:<line>] Task "<name>" handles secret data (parameter: <param>) but is missing no_log: true
Deprecated include:Use of bare include: (removed in ansible-core 2.12)[tasks:<line>] Deprecated: 'include:' has been removed. Use 'ansible.builtin.include_tasks:' or 'ansible.builtin.import_tasks:'

WARNING checks

CheckWhat to look forMessage format
Untagged tasksAny task without a tags: block[tasks:<line>] Task "<name>" has no tags — add at minimum component name + action category
Shell/command without idempotencyansible.builtin.shell or ansible.builtin.command without creates:, removes:, or changed_when:[tasks:<line>] Task "<name>" uses shell/command without idempotency guard (creates/removes/changed_when)
Both play-level and task-level becomebecome: true set at play level AND again on individual tasks[play:<name>] 'become: true' is set at play level; remove redundant task-level become declarations
Missing gather_facts declarationPlay with no explicit gather_facts:[play:<name>] gather_facts not explicitly set — add 'gather_facts: true' or 'gather_facts: false'
Deprecated include_role without name:ansible.builtin.include_role without explicit name: key[tasks:<line>] include_role missing explicit 'name:' key
ignore_errors: true without justificationignore_errors: true without adjacent comment explaining why[tasks:<line>] Task "<name>" uses ignore_errors: true without explanation — add a comment justifying this

INFO checks

CheckWhat to look forMessage format
Missing playbook headerNo YAML comment header with author/versionPlaybook is missing a documentation header (author, version, description)
No pre_tasks assertionFirst play has no version/OS assertionConsider adding an ansible.builtin.assert in pre_tasks to verify minimum Ansible version or OS compatibility
No post_tasks validationPlay has roles but no validation post_taskConsider adding a validation post_task (wait_for, uri) to confirm deployment success
No documented validation pathExample-style playbook has no clear validation task or surrounding test guidanceConsider adding validation tasks or README guidance aligned with references/testing.md
Long play without serialPlay targeting more than one group without serial:Play '<name>' targets multiple host groups without serial — consider serial: 1 for rolling updates
Hard-coded IP addressesAny IP address literal in the playbook[tasks:<line>] Hard-coded IP address '{{ ip }}' — use inventory variables or DNS names instead

Example Output

## Playbook Review: deploy-app.yml
Reviewed: 2026-03-31T10:00:00
Plays: 2  Tasks: 8  Roles: 2

---

### CRITICAL
- [play:Deploy application, tasks:line 15] Task "Restart app service" uses bare module name 'service' — must use FQCN 'ansible.builtin.service'
- [play:Deploy application, tasks:line 22] Task "Set admin password" handles secret data (parameter: password) but is missing no_log: true

### WARNING
- [play:Deploy application, tasks:line 18] Task "Run migration script" uses ansible.builtin.shell without idempotency guard (creates/removes/changed_when)
- [play:Deploy application, tasks:line 31] Task "Copy config file" has no tags — add at minimum component name + action category

### INFO
- Playbook is missing a documentation header (author, version, description)
- Consider adding a validation post_task (wait_for, uri) to confirm deployment success

---
Summary: 2 critical, 2 warnings, 2 info

Constraints

  • Never modify files. This command is read-only.
  • Do not suggest sed or awk commands to fix issues.
  • Do not generate "fixed" versions of the playbook.
  • Conclude with: "Use /ansible-designer:update-playbook to apply fixes."

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.