PentestGPTUnclaimed profile

Open-source autonomous pentesting agent: point the CLI at a target and it runs recon, exploitation, and reporting end-to-end.

Task: Coding & DevelopmentUse case: Autonomous penetration testing from reconnaissance to exploitationModels: gpt-3.5, gpt-4

0.0

based on 0 reviews

Are you the vendor?

reviews

Manzia generates an at-a-glance sentiment summary once an Agent has 5 approved reviews. This Agent has 0.

Summarized from 3 community tutorials · July 2026

Access: Free
API key: Bring your own
Runs on: CLI · Self-hosted
You’ll need: Git
Docker
Python 3 + pip
OpenAI or Anthropic API key
Linux VM (Kali/Ubuntu)
Skill level: Developer — code or a terminal

Summarized from 3 community tutorials · July 2026

Chains an LLM to drive pentest stepsConnects to a ChatGPT or Claude API/session and uses the model's reasoning to decide recon and exploitation commands against a target.
Generates a pentest procedure reportAfter a run, outputs a structured breakdown summarizing the penetration-test steps it executed.
Autonomously exploited a CTF box end-to-endFound a PHP file-upload vulnerability, got a reverse shell, escalated via a SUID Python binary, and retrieved user and root flags with minimal prompting.
Supports multiple LLM backendsWorks with an Anthropic API key, a Claude subscription login, OpenRouter, or a local LLM via LM Studio.
CLI command menu for guided operationShips CLI commands (e.g. help) to explore available pentest actions once connected to the model.

Summarized from 3 community tutorials · July 2026