AIRecon

AIRecon automates penetration testing and bug bounty reconnaissance for security testers. It guides an assessment through reconnaissance, analysis, exploitation and reporting, with a terminal interface for following the agent's work.

• Security testing tools: The agent uses a Kali Linux Docker sandbox, browser automation, Schemathesis API fuzzing and Semgrep code analysis during assessments.
• Caido integration: It can inspect and replay proxy traffic, automate fuzzing, and work with Caido findings, scope and sitemaps.
• Scope controls and audit logs: You can restrict allowed targets, block out-of-scope activity and keep a record of commands and requests.
• Persistent assessment memory: Resumable sessions retain chat and tool calls, while local memory uses verified or high-confidence findings and past tool results to guide later assessments.
• Security knowledge and extensions: An optional local knowledge base lets the agent search CVEs, testing techniques and payloads; MCP support connects it to additional tools.

AIRecon is open source under the MIT license and requires Docker and an OpenAI-compatible model gateway. Local inference through vLLM or a gateway connected to Ollama can operate without cloud services or API keys, keeping assessment data on your machine. Hosted GPT, Claude and Gemini models are also supported through compatible endpoints; those requests go to the external provider.

The model must support native tool calling. Smaller local models can handle simpler tasks but are more prone to tool-call errors and hallucinations.