Heads up This site is currently under heavy development.
← all tools
◆ AI/LLM Security

open-kritt

v1.4.0 open-source

Open-source, self-hosted AI vulnerability research tool that orchestrates agents to find and validate security issues in code.

Summary

Open-source, self-hosted AI vulnerability research tool that orchestrates agents to find and validate security issues in code.

Release history

  1. v1.4.0 Aug 12, 2026 · issue 011

    open-kritt v1.4.0 adds a headless runner, scan findings export, steps binding, and memory-aware engine controls.

    └──▷ GET THIS VERSION
    $ git clone --branch v1.4.0 https://github.com/Kritt-ai/open-kritt.git
    # already have the repo? check out this version:
    $ git checkout v1.4.0
    • Introduces the kritt-headless runner for executing scans on remote machines without a UI.
    • Adds export of completed scan findings, enabling practitioners to retrieve scan results for offline analysis or reporting.
    • Adds memory-aware runner admission and controls to the engine, preventing runners from being scheduled when memory thresholds would be exceeded.
    • Adds deterministic resume ordering controls to the engine, giving operators predictable restart behavior across paused scans.
    • Adds community star support to the frontend UI.
  2. v1.3.0 Aug 4, 2026 · issue 011

    open-kritt v1.3.0 adds ENGINE_IGNORE_LOW_STORAGE, separate post-processing models, scaled provider accounts, and a privacy-safe sharing loop.

    └──▷ GET THIS VERSION
    $ git clone --branch v1.3.0 https://github.com/Kritt-ai/open-kritt.git
    # already have the repo? check out this version:
    $ git checkout v1.3.0
    └──▷ TRY IT
    Allow the scan engine to continue running without halting when host disk space runs low.
    $ ENGINE_IGNORE_LOW_STORAGE=true
    • Adds ENGINE_IGNORE_LOW_STORAGE advanced setting for deployments where host disk usage is not a concern.
    • Supports separate post-processing models, allowing different models to be configured for the post-processing stage independently of the primary scan model.
    • Scales provider accounts and scan processing to handle larger workloads.
    • Hardens scan runtime and account handling for more robust operation.
    • Adds a privacy-safe sharing loop in the frontend.
    +2 moreshow less
    • Displays the active worker model and harness in the UI for better operational visibility.
    • Improves active worker status display in the frontend.
my-toolchain — 0 tools
paste an install list to detect your tools

A brew list, a Brewfile, requirements.txt, a Dockerfile — or just the product names, free-form. Nothing leaves your browser.

    browse all tools →