Repolode
Submit
zakirkun

zakirkun/deep-eye

AI-driven penetration testing tool that uses multiple LLM providers to generate payloads, scan web targets for vulnerabilities, and write compliance-mapped reports.

Open-source alternative to:Burp Suite Professional

View on GitHub
Stars
2,341
Forks
424
Open issues
13
Last commit
14 days ago

What is it?

Deep Eye is a Python command-line penetration testing tool for web targets. It routes payload generation and triage through a set of AI providers, including OpenAI, Claude, Gemini, Groq, Mistral, and local models via OLLAMA or LM Studio, and runs checks for more than 50 vulnerability types such as SQL injection, XSS, SSRF, JWT weaknesses, IDOR, GraphQL issues, and CORS/CSP misconfigurations.

Beyond scanning, it generates findings with per-vulnerability evidence summaries, filters false positives, deduplicates results by fingerprint, and exports reports in HTML, PDF, JSON, SARIF, JUnit, CSV, and XLSX. Findings can be mapped to PCI-DSS v4, SOC2 CC, and ISO 27001:2022 controls. Scan diffs and a retest mode keep only findings that are new compared with a baseline.

The tool also includes OpenAPI-based crawling, Playwright browser automation, a mitmproxy integration, CAPTCHA detection, multi-role authentication helpers, and Nuclei-style YAML templates. It is intended only for authorized security testing.

Key features

  • 01Failover across 11 AI providers through a single generate() abstraction.
  • 02Scans for 50+ vulnerability types, including SQLi, XSS, SSRF, JWT, IDOR, and GraphQL.
  • 03Seeds crawling from OpenAPI or Swagger specs.
  • 04Exports reports in HTML, PDF, JSON, SARIF, JUnit, CSV, and XLSX.
  • 05Maps findings to PCI-DSS v4, SOC2 CC, and ISO 27001:2022.
  • 06Compares scans against a baseline and retests only new findings.

Use cases

  • Running authorized web application security assessments with AI-assisted payload generation.
  • Producing compliance-oriented reports for PCI-DSS, SOC2, or ISO 27001 audits.
  • Integrating scan results into CI pipelines via SARIF or JUnit output.
  • Tracking regressions by diffing scans against a saved baseline.

Quick start

  1. 1

    Clone the repository and enter its folder

  2. 2

    Install dependencies on Linux/Mac with the install script, then activate the virtualenv

    chmod +x scripts/install.sh && ./scripts/install.sh && source .deep-venv/bin/activate
  3. 3

    Run a basic scan against a target you are authorized to test

    python deep_eye.py -u https://target.com

Things to know

  • Requires at least one AI provider API key or a local OLLAMA setup before it can run.
  • Only use it on systems you own or have explicit written permission to test.
  • The repository metadata lists the license as NOASSERTION, although the README states MIT.

Topics

#penetration-testing#security-scanner#llm#vulnerability-scanner#web-security#bug-bounty#python#compliance

FAQ

What is deep-eye?
AI-driven penetration testing tool that uses multiple LLM providers to generate payloads, scan web targets for vulnerabilities, and write compliance-mapped reports. Deep Eye is a Python command-line penetration testing tool for web targets. It routes payload generation and triage through a set of AI providers, including OpenAI, Claude, Gemini, Groq, Mistral, and local models via OLLAMA or LM Studio, and runs checks for more than 50 vulnerability types such as SQL injection, XSS, SSRF, JWT weaknesses, IDOR, GraphQL issues, and CORS/CSP misconfigurations.
Is deep-eye free to use? What license does it have?
deep-eye is open source under the Other license. Check the license terms before using it in commercial products.
Is deep-eye still maintained?
Actively maintained. The last commit to deep-eye was on Oct 6, 2026.
What can I use deep-eye for?
Running authorized web application security assessments with AI-assisted payload generation. Producing compliance-oriented reports for PCI-DSS, SOC2, or ISO 27001 audits. Integrating scan results into CI pipelines via SARIF or JUnit output. Tracking regressions by diffing scans against a saved baseline.
How do I install deep-eye?
Clone the repository and enter its folder Install dependencies on Linux/Mac with the install script, then activate the virtualenv: chmod +x scripts/install.sh && ./scripts/install.sh && source .deep-venv/bin/activate Run a basic scan against a target you are authorized to test: python deep_eye.py -u https://target.com
Where can I get deep-eye?
The source code and installation instructions are on GitHub: https://github.com/zakirkun/deep-eye

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects