Repo Reality Check
Paste any public GitHub repository and watch five inspector agents read it live: how much merged code was never reviewed, which dependencies float unpinned, whether the pipeline has any gates, and which guardrail files are missing. Findings come from the repository itself, not a questionnaire.
Scan a repository NewAI Coding Risk Audit
Your developers are already using AI to write code. Eight plain questions reveal how exposed that leaves you — with a scored report, your biggest gaps, and a starter policy to hand your team. Written for non-technical readers.
Take the audit InteractiveHarness Builder
Generate a drop-in rules file (AGENTS.md / CLAUDE.md) that turns a sycophantic coding agent into an expert. Toggle the behaviors you want, then copy or download. Runs entirely in your browser.
Open the builder InteractivePR Review Checklist Generator
Reviewing agent-written code? Pick your stack, risk lane and change type — get a paste-ready markdown checklist that hunts the failure signatures agent code actually has.
Generate a checklist → ComplianceEU AI Act Risk Classifier
Four questions to place your AI system in the right EU AI Act risk tier — prohibited, high-risk, limited, or minimal — and see the obligations that follow. Guidance, not legal advice.
Classify a systemMore tools in the works.