repowiserepowise

Command-line tool

elementalsouls / claude-bughunter

A Claude Code skill bundle for bug hunting and external red-team work - 82 skills, 15 slash commands, 681 disclosed-report patterns curated across 24 core vulnerability classes, plus enterprise identity + infrastructure attack matrices.

★ 15,850stars on GitHubPythonMIT

This connects Repowise to elementalsouls/claude-bughunter's code. It does not install claude-bughunter.

Repowise maps public repos from their source. Not affiliated with elementalsouls. What is Repowise?

github.com/elementalsouls/claude-bughunterIndexed at 0e70a02

What is claude-bughunter?

claude-bughunter is a Claude Code skill bundle for bug hunting and external red-team work, aimed at security researchers and red-team operators who test targets they are authorized to test. Each skill is a folder with a SKILL.md file that Claude Code loads automatically when a task matches the skill's topic, so no skill has to be invoked by name. The bundle also includes a command-line tool and a scoped test engine, and its main language is Python. To install it, run git clone https://github.com/elementalsouls/claude-bughunter.git and follow the steps below. Its code health is 7.2 out of 10 (Good), ranked #123 of 4,000+ indexed repos. The last commit to claude-bughunter was on Nov 18, 2026, and 18 people have contributed.

As of commit 0e70a02, indexed Nov 20, 2026. What it is comes from a model reading the README; the rest is measured from the source and git history, not from the maintainer.

For agentsclaude-bughunter.md

Repo facts: type, license, health, rank, contributors
claude-bughunter facts
TypeCommand-line tool
LicenseMIT
Code health7.2 / 10 (Good)
Health rank#123 of 4,000+ indexed repos
Last commitNov 18, 2026
Contributors18
Bus factor2
Size308 files, 10,692 lines

Install claude-bughunter

Commands copied from the README

  1. 1Clone the claude-bughunter repository.
    git clone https://github.com/elementalsouls/claude-bughunter.git
  2. 2Move into the claude-bughunter folder.
    cd claude-bughunter
  3. 3Inside Claude Code, add the claude-bughunter marketplace as a plugin source.
    /plugin marketplace add elementalsouls/Claude-BugHunter
  4. 4Install the claude-bughunter plugin from that marketplace.
    /plugin install claude-bughunter@elementalsouls
  5. 5Alternatively, from a clone of the repository on macOS or Linux, run the copy installer to copy the skills and commands into the Claude configuration folder.
    bash scripts/install.sh

Releases on GitHub (latest v2.1, Jun 5, 2026)

Full install notes on GitHub

Explore claude-bughunter

Ask about this code in your agent

This connects Repowise to claude-bughunter's code. It does not install claude-bughunter. To install claude-bughunter, use the install steps.

Useful if you want to change or fork claude-bughunter: your agent can ask how it is put together and get answers that cite its files.

Ask about claude-bughunter

Answers come from claude-bughunter's README and source, with the files they cite. Two free answers a day without an account.

Featured answer

claude-bughunter is a Claude Code skill bundle for bug hunting and external red-team work, aimed at security researchers and red-team operators who test targets they are authorized to test, and it contains 82 skills and 15 slash commands across 308 files. Each skill is a folder holding a SKILL.md file that Claude Code loads automatically when a task matches the skill's topic, and the skills folder holds one such folder per skill. Shell commands enter through cbh/cli.py in the CLI layer, while engine/engine.py runs a scoped test of one target using a scope file of in-scope and out-of-scope hosts, and the README says the engine routes each finding to the matching skill through engine/skill_map.py. The scripts directory creates engagement folders, installs the skill bundles, and defines the hunt shell function in scripts/hunt.sh, and eval/run_eval.py measures whether the hunt skills help an autonomous agent find and exploit web bugs. To start reading, open cbh/cli.py for the entry point, engine/engine.py for the test flow, and one SKILL.md for how a skill is written; by file count the layers are CLI 15, Application 144, Data 1, Config 18, and Docs & Tooling 88.

Files per layer in claude-bughunter5 layers.

Basis: Inferred · knowledge graph· as of 0e70a02

Want this for your own codebase? Map your repo free

About this page

Get a page like this for your own codebase

Repowise built this map of elementalsouls/claude-bughunter from its source and git history. It parses every symbol, draws the dependency graph, scores each file's health from complexity, duplication, coverage and churn, finds hotspots and owners in the history, and writes documentation for every module. Your coding agent reads the same index over MCP, so its answers cite real files.

Free for 2 public repos, with a page like this one. No card. See how Repowise works

Prefer local? pip install repowise && repowise init indexes your own checkout, no account needed. Star Repowise on GitHub

elementalsouls/claude-bughunter FAQ

What is claude-bughunter?

claude-bughunter is a Claude Code skill bundle for bug hunting and external red-team work. It includes 83 skills and 15 slash commands, covering web app vulnerability classes and enterprise platform attack matrices.

How does claude-bughunter work?

claude-bughunter loads each skill automatically when a task matches its topic, so describing what you are testing is enough. Its scoped engine maps a target's attack surface from a scope file that lists in-scope and out-of-scope hosts.

How do I install claude-bughunter?

claude-bughunter installs as a Claude Code plugin from inside Claude Code, using the marketplace and install commands. A copy install from a clone is also available through the install script on macOS and Linux. See the install steps

What language is claude-bughunter written in?

claude-bughunter is written mainly in Python, with 88 Python files. It also includes 4 shell files and 2 Dockerfiles.

How is the claude-bughunter code organized?

claude-bughunter keeps its skills in the skills folder and its command-line package in cbh. The engine folder holds scoped testing, scripts holds installers and engagement scaffolding, and eval measures the hunt skills.

Who is claude-bughunter for?

claude-bughunter is for security researchers and red-team operators who test targets they are authorized to test. It covers the external attack surface, meaning anything reachable from the internet.

What license does claude-bughunter use?

claude-bughunter is released under the license with SPDX identifier MIT. The license is stated in the repository metadata.

Code health

7.1out of 10Good

Hotspots average 4.4, the place fixes pay off most.

On the leaderboards:#31 of 67 Python repos

Key features of claude-bughunter

  • Skills load automatically by topic, so the user describes what they are testing in plain English and no skill has to be invoked by name.
  • The bundle includes 83 skills and 15 slash commands, including /hunt, /recon and /report.
  • Hunt skills cover 24 core vulnerability classes, with patterns curated from 681 disclosed HackerOne reports.
  • Enterprise platform attack matrices cover identity and infrastructure targets such as M365/Entra, Okta and vCenter.
  • A scoped engine maps a target's attack surface from a scope file and routes each finding to the matching skill.

How elementalsouls/claude-bughunter works

Commands typed at the shell enter through cbh/cli.py, the entry point that parses a typed command and runs it. The engine directory holds engine/engine.py, which runs a scoped security test of one target using a scope file that lists in-scope and out-of-scope hosts. The README says the engine routes each finding to the skill that handles it, and engine/skill_map.py is one of the codebase's hotspot files. The skills folder holds one folder per skill, each with a SKILL.md file that Claude Code loads when a task matches its topic. The scripts directory creates engagement folders, installs the skill bundles, and defines the hunt shell function in scripts/hunt.sh. The eval module includes eval/run_eval.py, which measures whether the hunt skills help an autonomous agent find and exploit web bugs.

The codebase has 308 files and 727 symbols in 10 modules, led by Python, Shell and Dockerfile. Code health is 7.1 out of 10, rated good.