open-kritt: an AI agent orchestration platform for finding and validating security vulnerabilities in code

What it solves

Finding vulnerabilities in large codebases is difficult for AI models when they are asked to analyze the entire repository at once. open·kritt solves this by breaking security research into small, focused tasks that can be executed by AI agents in parallel, reducing noise and improving the accuracy of findings.

How it works

The platform allows users to build "playbooks" or workflows by chaining together focused prompts. These agents run in disposable Docker containers with root access and internet connectivity, enabling them to install tools, compile code, and build proofs of concept. The system then aggregates the results, applying automatic de-duplication and custom severity rankers to prioritize the most critical issues.

Who it’s for

It is designed for security researchers and security-minded developers who want full control over their prompts, workflows, model providers (such as OpenAI, Anthropic, or OpenRouter), and infrastructure.

Highlights

  • Workflow Builder: Create reusable security research playbooks by chaining prompts.
  • Agentic Execution: Agents can run tests and compile targets in isolated environments to verify findings.
  • Flexible Model Access: Supports multiple providers including Codex, OpenAI, Anthropic, and OpenRouter.
  • Automated Prioritization: Features a consistent finding schema and automatic de-duplication to rank vulnerabilities.

Sources