agentic-actions-auditor

Audits GitHub Actions workflows for security vulnerabilities in AI agent integrations including Claude Code Action, Gemini CLI, OpenAI Codex, and GitHub AI Inference. Detects attack vectors where attacker-controlled input reaches. AI agents running in CI/CD pipelines.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a well-structured, highly actionable security audit skill with clear methodology sequencing and appropriate progressive disclosure to reference files. Its main weakness is verbosity — the 'Rationalizations to Reject' section, extensive 'When to Use/NOT to Use' lists, and explanatory prose about why certain patterns are dangerous add token overhead that could be trimmed since Claude can infer these rationales. The workflow clarity and actionability are excellent, with concrete commands, precise field names, and structured output formats.

Suggestions

Compress the 'Rationalizations to Reject' section into a compact table (rationalization | why wrong) rather than paragraph-per-item format to save ~40% of that section's tokens.

Trim the 'When to Use' and 'When NOT to Use' sections — Claude can infer appropriate usage from the methodology itself; consider reducing to 2-3 bullet points each or removing entirely.

Dimension	Reasoning	Score
Conciseness	The skill is thorough but verbose in places. The 'Rationalizations to Reject' section, 'When to Use'/'When NOT to Use' lists, and explanatory prose add significant token overhead. Some sections like URL parsing and bash safety rules are well-structured but could be more compact. The skill explains concepts Claude would understand (e.g., what pull_request_target does, what sandboxes are) rather than just stating the rules.	2 / 3
Actionability	The skill provides highly concrete, executable guidance: specific `gh api` commands with exact flags and jq expressions, precise YAML field names to check per action type, a structured detection methodology with specific pattern-matching rules, and detailed report formatting instructions. The action reference table, security context capture fields, and vector detection heuristics are all specific and actionable.	3 / 3
Workflow Clarity	The 5-step methodology (Discover → Identify → Capture → Analyze → Report) is clearly sequenced with each step building on the previous. Step 0 handles mode determination with error handling. Cross-file resolution has a depth limit. The reporting step includes severity judgment criteria, data flow trace rules, and clean-repo output handling. Validation checkpoints are present (e.g., 'If no workflow files found, stop'; 'If no AI action steps found, stop').	3 / 3
Progressive Disclosure	The skill maintains a clear overview structure with well-signaled one-level-deep references to detailed materials: action-profiles.md, foundations.md, individual vector files (vector-a through vector-i), and cross-file-resolution.md. The main SKILL.md provides the methodology overview and quick-check table while deferring detection heuristics and action-specific details to reference files. Navigation is clear via the vector table and 'Detailed References' section.	3 / 3
	Total	11 / 12 Passed

Description

82%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description excels at specificity and distinctiveness, clearly naming the tools, domain (GitHub Actions security), and attack patterns it addresses. Its main weakness is the lack of an explicit 'Use when...' clause, which caps completeness at 2. Minor formatting issues with extra spaces exist but don't significantly impact clarity.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when the user asks to review or audit GitHub Actions workflows for security issues, especially those involving AI agents or code generation tools in CI/CD pipelines.'

Consider adding trigger terms like 'workflow security', 'prompt injection', 'CI security audit', or 'supply chain attack' to capture more natural user queries.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: audits GitHub Actions workflows, detects security vulnerabilities in AI agent integrations, names specific tools (Claude Code Action, Gemini CLI, OpenAI Codex, GitHub AI Inference), and identifies specific attack vectors involving attacker-controlled input reaching AI agents in CI/CD pipelines.	3 / 3
Completeness	The 'what' is well-covered (audits workflows for security vulnerabilities, detects attack vectors), but there is no explicit 'Use when...' clause or equivalent trigger guidance telling Claude when to select this skill. The 'when' is only implied by the description of capabilities.	2 / 3
Trigger Term Quality	Includes strong natural keywords users would say: 'GitHub Actions', 'security vulnerabilities', 'AI agent', 'CI/CD pipelines', 'Claude Code Action', 'Gemini CLI', 'OpenAI Codex', 'attack vectors'. These are terms a user concerned about workflow security would naturally use.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive niche: specifically targets security auditing of GitHub Actions workflows for AI agent integrations. The combination of CI/CD security + AI agents + named tools makes it very unlikely to conflict with other skills.	3 / 3
	Total	11 / 12 Passed

Validation

90%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: popey/claude-code-skills
Path: skills/agentic-actions-auditor/SKILL.md
Commit: a5a6601

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.