Claude Plugin
Last commit on August 5, 2026·Created on January 14, 2026
trailofbits/skills
“Security auditing for AI agent integrations within GitHub Actions workflows”
Combined rank
#50
across all skills
In Security
#4
category rank
Stars
6.4k
+1.8% in last 7d
Forks
556
+2.0% in last 7d
Watchers
61
-6.2% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
Description
This skill identifies security vulnerabilities in CI/CD pipelines that integrate AI agents such as Claude Code Action, Gemini CLI, and OpenAI Codex. It helps security researchers detect attack vectors where attacker-controlled input reaches AI agents through environment variables, expression injection, or permissive sandbox configurations.
By analyzing workflow files for prompt injection risks and dangerous action configurations, the tool provides a structured way to evaluate the safety of agentic automation in deployment pipelines.
WHO IT'S FOR
Security researchers
performing deep architectural code analysis
Smart contract auditors
scanning Algorand contracts for vulnerabilities
DevSecOps engineers
auditing CI/CD pipelines for AI agent risks
Technical leads
preparing codebases for formal security reviews
30 Days of GitHub Stars
+426 stars↑ 7.2% monthly change
Jul 7Jul 14Jul 21Jul 28Aug 5
Compatibility & Install77 Skills · 4 Compatible agents · 3 categories · 1 install path
Compatible agents
❯Claude Code❯Codex❯Cursor❯Gemini CLI
Categories
SecurityResearchSoftware Engineering
Install via
Repository instructions
Also in this category4 of 24