# gadievron/raptor

Raptor turns Claude Code into a general-purpose AI offensive/defensive security agent. By using Claude.md and creating rules, sub-agents, and skills, and orchestrating security tool usage, we configure the agent for adversarial thinking, and perform research or attack/defense operations.

Repository: https://github.com/gadievron/raptor
Canonical: https://ross.abutalabs.com/products/raptor
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-08-26T12:17:28+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 48, longevity 22
- inputs: {"age_days": 320, "days_push": 7, "days_rel": 132, "gap_med": null, "n_releases_24m": 1}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3672, forks 569 (observed 2026-08-28T04:08:13.521845+00:00)

## What it is
RAPTOR is an autonomous offensive/defensive security research framework built on top of Claude Code, chaining static analysis, binary analysis, LLM-powered vulnerability validation, exploit generation, and patch writing into a single workflow. It configures the AI agent for adversarial thinking via rules, sub-agents, and skills, and orchestrates security tool usage against codebases or binaries.

## Use cases
- run autonomous penetration testing against a codebase
- validate and triage vulnerabilities with an LLM agent
- generate exploits for discovered bugs
- automate security research on binaries
- configure Claude Code as a red team / blue team agent
- automatically write patches for found vulnerabilities

## When to choose
- you already use Claude Code and want a security-focused agent setup
- you need an orchestrated pipeline from static/binary analysis to exploit and patch generation
- you want an extensible framework with sub-agents and skills for adversarial security research

## When to avoid
- you need polished, production-grade security tooling with support guarantees
- you cannot use Claude Code or an equivalent LLM analysis layer
- your use case is commercial and incompatible with the CodeQL license restrictions

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, security, penetration-testing, vulnerability-scanning, llm-inference, prompt-engineering
- domain: security, penetration-testing, artificial-intelligence, developer-tools
- platform: python, cli
- tags: claude-code, offensive-security, vulnerability-research, exploit-development, red-team, autonomous-agents, binary-analysis, static-analysis, ai-agents, linux, macos

## Member repositories
- gadievron/raptor (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:13.521845+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:31:18.401893+00:00, confidence not recorded.
  - readme: https://github.com/gadievron/raptor (fetched 2026-08-28T04:08:13.521845+00:00, sha 548b09911e0b)
- Data as of 2026-08-30T08:39:29.467469+00:00.