
Kritt
✓ Text source-checked Nov 16, 2026Kritt breaks a codebase into focused security tasks, runs AI agents in parallel, and turns their work into ranked, verified findings.
Source-checked against the vendor's site · rebuilt from the vendor's own pagesProfile updated Link checked Pricing captured How we rate
In short
Kritt uses parallel AI agents to break codebases into security tasks, producing ranked, verified findings. Choose free self-hosting or paid managed scans.
Kritt is an AI-assisted code security research platform that breaks a codebase into focused security tasks and runs AI agents in parallel to produce ranked, verified findings. Users can self-host the open-source engine or commission the Kritt research team to run a managed scan.
The open·kritt component is an open-source distillation of the internal workflow that allows users to inspect and extend the ranked findings. Verification passes run post-scripts per finding to validate exploitability, build a proof-of-concept, or draft a report. The system ranks, normalizes, and de-duplicates findings against user-defined impact criteria.
The platform was shaped by workflows used in public bug-bounty research.
Enterprise engagements provide tailored security research for teams requiring CI/CD integration, release coverage, custom volumes, and SLAs.
According to Kritt: Kritt is an AI-assisted code security research platform that breaks a codebase into focused security tasks and runs AI agents in parallel to produce ranked, Verified findings. Its pages list this strength: The platform runs AI agents in parallel to process security tasks.
— The AI Tool Bible editorial team
Pros
- ✅ The platform runs AI agents in parallel to process security tasks.
- ✅ The output consists of ranked and verified findings.
- ✅ Self-hosting allows users to choose their own models, workflows, repositories, and severity rules.
- ✅ Post-scripts can be run to validate exploitability, build a proof-of-concept, or draft a report.
Pricing (captured )
| Tier | Monthly | Key features |
|---|---|---|
| Single Scan | $5,000 per scan |
|
| Enterprise | Contact us |
|
Prices change; we show the grid as captured on Aug 11, 2026. Check current pricing on the vendor's site →
Facts
- Platforms
- linux, macos, windows
- Free trial
- Yes (not source-checked)
- Open source
- Yes
- Model used
- Codex and Claude Code (as agent harnesses) (not source-checked)
- Hacker News
- 1 story · top 2 points · last Jan 26, 2026 · checked Nov 14, 2026
- Earliest Wayback capture
- 2026-07-27 (earliest archived capture of the site, not a launch date)
Where this tool sits
Use cases
Frequently asked
- How much does Kritt cost?
- Kritt offers a freemium model. The open-source engine (open·kritt) is free to self-host. Managed single scans cost $5,000, with enterprise engagements available for custom volumes and SLAs.
- What AI models power Kritt?
- Kritt uses Codex and Claude Code as agent harnesses. These models run in parallel to execute focused security tasks derived from your codebase, helping generate and verify findings.
- Can I self-host Kritt?
- Yes, the open·kritt component is an open-source distillation of the internal workflow. It allows users to self-host, inspect, and extend the ranked findings, including running verification passes and post-scripts.
- How does Kritt verify findings?
- Verification passes run post-scripts per finding to validate exploitability, build a proof-of-concept, or draft a report. The system then ranks, normalizes, and de-duplicates findings against user-defined impact criteria.
- Is Kritt suitable for continuous security coverage?
- Yes, Kritt is designed for continuous security coverage and public bug-bounty research. Enterprise engagements provide tailored research for teams requiring CI/CD integration, release coverage, and custom SLAs.