agentic-actions-auditor

Audit GitHub Actions workflows for prompt-injection and data-exfiltration risks.

31|4|Updated Apr 6, 2026

Search Tags:#github-actions #ai-security #prompt-injection #workflow-audit #ci-cd-security #cross-file-resolution

Ready to run

npx skills add https://github.com/kissrosecicd-hub/agents-evolution --skill agentic-actions-auditor-kissrosecicd-hub

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agentic-actions-auditor
Source: https://github.com/kissrosecicd-hub/agents-evolution/tree/main/.agents/skills/tob-agentic-actions-auditor
Command: npx skills add https://github.com/kissrosecicd-hub/agents-evolution --skill agentic-actions-auditor-kissrosecicd-hub

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audits GitHub Actions workflows for AI agent security vulnerabilities, focusing on prompt injection and data-exfiltration risks in agent integrations.

Core Features & Use Cases

Identify AI agent actions in workflows and map how prompts and system prompts are formed.
Trace attacker-controlled inputs from events, env vars, and workflow inputs to prompt content.
Produce actionable findings with remediation guidance and cross-file risk patterns.

Quick Start

Audit a repository's AI-action workflows for prompt-injection risks with a single workflow containing an AI action.

Frequently Asked Questions about agentic-actions-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I audit GitHub Actions workflows for prompt injection risks?▼

Auditing GitHub Actions workflows for prompt injection involves identifying AI action steps and tracing attacker-controlled inputs from events or env vars to prompt content, producing findings with remediation guidance.

What AI agent integrations are vulnerable to data exfiltration in CI/CD pipelines?▼

AI agent integrations using Claude Code Action, Gemini CLI, OpenAI Codex, or GitHub AI Inference are vulnerable to data exfiltration when untrusted inputs reach prompts via push, pull_request_target, or workflow_call triggers.

How does cross-file resolution detect security risks in nested GitHub workflows?▼

Cross-file resolution detects security risks in nested GitHub workflows by analyzing cross-file references and capturing security-relevant inputs from with: blocks and env: blocks across the triggered workflow chain.

Can I trace workflow inputs to AI system prompts in pull_request_target events?▼

Yes, you can trace workflow inputs to AI system prompts in pull_request_target events by capturing security-relevant inputs from with: blocks and env: blocks, mapping how prompts and system prompts are formed.

What are the limitations when auditing AI agent actions in GitHub Actions?▼

Auditing limitations include handling deeply nested cross-file references in complex workflow_call chains, where tracing attacker-controlled inputs to prompt content requires complete visibility into env: and with: blocks.