agentleFS
noamgariani11noamgariani11/claude-skills/skills/vuln-research/SKILL.md

vuln-research

Skill0 starsChanged 4 months agoNo licenceView on GitHub

The white-box sibling of /bug-hunt ("bug-hunt part two") — find reportable vulns by READING CODE, reverse-engineering CLIENTS, and OFFLINE-FUZZING parsers when an in-scope bug-bounty program (HackerOne or BugCrowd) asset's source or binary is OBTAINABLE: an open-source product, a public SDK, a published npm/PyPI/Maven/ crates package, a downloadable desktop/Electron app, a mobile APK/IPA, a browser extension, or firmware — plus the OSS dependencies of a closed product. Where /bug-hunt probes the live host black-box, this analyzes the artifact white/grey-box and mostly OFFLINE: obtain & build it, run SAST + taint/sink-to-source (sast_scan.py: Semgrep + a dependency-free built-in sink scan), multiply one root cause into N with variant analysis (the disclosed-bug -> Semgrep/CodeQL-rule -> all sibling repos loop), decompile thick clients (unpack_app.py: asar/APK/JAR/.NET/crx/firmware -> source + secret/endpoint/sink mine), reverse native binaries (Ghidra/Frida/strings), and coverage-fuzz parsers on a LOCAL copy (AFL++/libFuzzer/boofuzz/radamsa/Jazzer + ASAN). Grounded in research/27-source-review-and-variant-analysis.md + research/28-binary-re-and-offline-fuzzing.md (distilled from "From Day Zero to Zero Day", Lim 2025). Shares /bug-hunt's workspace, guardrails, validator gate, report tooling, and journal; hard-codes the researcher handle monk11. Never fuzzes/ loads/crashes the live host (that's DoS); never auto-submits; every claim grounded in a real path:line or a crash you captured (anti-slop). Use when: "vuln-research", "/vuln-research", "review this source for bugs", "the target is open source / ships an SDK / has a desktop or mobile app", "decompile and audit this client", "fuzz this parser", "find a CVE in this dependency", "white-box hunt", "bug-hunt part two", "bugcrowd vuln research", "bugcrowd white-box". NOT for black-box live-host hunting (use /bug-hunt) or auditing your OWN app (use /security-dude).

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. 0 are common, 8 are distinctive, and 10 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 10

    Very specific

    Tied to their project's names, files or tools. Probably skip.

No licence file, so all rights are reserved — read it at the source. We count its rules but don't quote them. Read them on GitHub.

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.