About this project
hackingtool is a Python-based CLI toolkit that consolidates 215 curated security testing tools into one console, organized across 21 categories including information gathering, web attacks, wireless attack, phishing, SQL injection, forensics, post-exploitation, password cracking, and more.
The distinguishing feature is its AI layer: users can describe a goal in plain English (e.g., "find subdomains of example.com") and the tool maps that intent to the right tools with the exact documented command. The `/goal` command drafts a step-by-step plan, asks for authorization confirmation, then runs each step via list-form subprocess calls — never through a shell. Tool output is not fed back to the model after planning.
`/find` searches the local catalog first, then the GitHub API, ranking results with explanations and never installing or running anything automatically. Out-of-scope requests (jamming, DoS, mass-targeting) are refused before any network call.
Other notable features include background execution via tmux panes (`/run ... &`), a headless mode for scripted engagements producing normalized `findings.json` and Markdown reports, tag-based discovery across a fixed 63-tag taxonomy, and safe install practices (no `curl | bash`, SHA-256 verified, signed releases with SBOM).
The AI layer is opt-in and bring-your-own-key — supporting OpenAI-compatible endpoints or a local Ollama instance — and degrades to deterministic offline behavior when no model is configured.
Requirements: Python 3.10+, Linux or macOS. Windows is not supported. Optional runtimes include Go (for tools like nuclei, ffuf, amass), Ruby, tmux, and Docker.
Installation is via pipx (recommended), uv, a plain venv, or Docker. The project emphasizes authorized security testing only and provides extensive documentation on safe operation.
Comments
0 Rating appears after 10 ratings
Sign in to join the discussion.