hunt-rce — Claude Code skill: what it does, how to install
hunt-rce — a skill for hunting remote code execution bugs: it points to likely RCE spots across web apps, CI/CD, containers and ML services, with ready grep signatures.
no license ★ 1,298 updated 2026-06-12
npx skills add H-mmer/pentest-agents --skill hunt-rce -a claude-codeWhat hunt-rce does
The skill draws on 1,218 public RCE reports from HackerOne, Project Zero, Intigriti and GitHub Security Advisories, with recent entries cross-checked against NVD. It walks through target classes: deserialization bugs in React Server Components, takeovers of CI/CD runners and GitOps controllers, container runtime escapes (runc, BuildKit, NVIDIA Container Toolkit), pickle-based deserialization in ML inference pipelines, and prompt injection into agentic LLM tools like PythonREPLTool. Each class comes with named CVEs, CVSS scores and disclosure sources.
A separate section lists grep signatures: rg patterns for spotting risky sinks such as ObjectInputStream/readObject in Java or pickle.loads/yaml.load in Python directly in a codebase. Treat it as a bounty-hunting checklist rather than a scanner — it's built for targets where user input can actually reach a shell, deserializer, template engine or git/curl/tar call.
The skill is written for bug bounty and penetration testing work within an authorized scope.
What it helps with
- spot an RCE path in a modern JS stack using React Server Components
- check a CI/CD runner or GitOps controller for a git-resolver takeover
- assess container escape risk through runc or BuildKit
- find unsafe pickle deserialization in an ML inference server
- catch prompt injection reaching agentic LLM tools like PythonREPLTool
Example prompts for the agent
- Run hunt-rce against this Next.js app to check for an RCE path through React Server Components
- Use hunt-rce to scan this Java service for risky deserialization signatures
- Apply hunt-rce to the CI pipeline to check the GitOps resolvers for argument injection
How to install hunt-rce
1Skills CLI (Vercel)
npx skills add H-mmer/pentest-agents --skill hunt-rce -a claude-code2Manually: put the files in place
~/.claude/skills/pentest-agents-hunt-rce/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/pentest-agents-hunt-rce/
1Skills CLI (Vercel)
npx skills add H-mmer/pentest-agents --skill hunt-rce -a codex2Manually: put the files in place
.agents/skills/pentest-agents-hunt-rce/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add H-mmer/pentest-agents --skill hunt-rce -a cursor2Manually: put the files in place
.cursor/skills/pentest-agents-hunt-rce/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install H-mmer/pentest-agents providers/cursor/.cursor/skills/pentest-agents-hunt-rce2Manually: put the files in place
.github/skills/pentest-agents-hunt-rce/SKILL.md1Skills CLI (Vercel)
npx skills add H-mmer/pentest-agents --skill hunt-rce -a gemini-cli2Platform CLI
gemini skills install https://github.com/H-mmer/pentest-agents --path providers/cursor/.cursor/skills/pentest-agents-hunt-rceManually: put the files in place
.devin/skills/pentest-agents-hunt-rce/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about hunt-rce
How do I install hunt-rce in Claude Code?
Run npx skills add H-mmer/pentest-agents --skill pentest-agents-hunt-rce -a claude-code; the skill lands at ~/.claude/skills/pentest-agents-hunt-rce/SKILL.md.
What does hunt-rce do?
It's a bug-hunting skill focused on remote code execution: it ranks target classes, from React Server Components to ML inference servers, and supplies grep signatures for finding risky deserialization sinks in code.
What research backs hunt-rce?
It's built on 1,218 public RCE reports plus a set of 2024-2026 CVEs verified against NVD, including React Server Components (CVE-2025-55182), runc Leaky Vessels (CVE-2024-21626), BentoML pickle (CVE-2025-27520/32375) and LangChain REPL (CVE-2025-68613).