PentestGPT

Autonomous penetration testing agent that drives Claude Code or Codex through staged recon, exploit and report pipelines.

Agent by Gelei Deng

IndexedMIT★ 16,099

Write a review

Get started

Setup steps are in the repository. Repository

About

PentestGPT is an open-source penetration testing agent published at USENIX Security 2024. Version 1.0 runs staged pipelines, from recon through exploit to walkthrough for CTF challenges, or from asset discovery through vulnerability identification to report for pentests, and drives Claude Code or Codex to run the tools. A legacy interactive mode supports several model providers, including local Ollama, and sessions can be saved and resumed.

Use cases: security, security-scanning

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Fact sheet

Jobs: securityREADME
Autonomy: Autonomous: runs tasks end to end and reports backREADME
Where you use it: TerminalREADME
Hosting: You run itREADME
Models: anthropic, openai, google, deepseek, xai, qwen, moonshot, ollamaREADME
Open source: YesREADME

Facts extracted by Claude Haiku 5.5 from the linked sources on Oct 24, 2026. Wrong fact? Tell us.

Built on

Frameworks named in its dependency files.

0.730score
Claude Agent SDK (Python)
Anthropic
Python SDK that runs the Claude Code agent loop inside your own program, with custom tools and hooks written as Python functions.
HarnessAFreshness: A★ 8.3K↓ 7.9M/wk

Signals

Stars: 16,099
Forks: 2,825
Open issues: 72
Last commit: Jul 14, 2026 (3 months ago)
Last release: Dec 24, 2025 (10 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/GreyDGL/PentestGPT
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: README github-graphql

Cite this page

PentestGPT. Agent Mart. https://agentmart.dev/agents/pentestgpt (last updated 2026-10-24)