Kaval runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 11 Sep 2026.
Verify the facts an agent's action depends on before it acts — ALLOW, REVIEW, or BLOCK.
What changed 41
Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 9 August 2026. No other catalogue keeps this.
Nothing serious here today
Today is the operative word: we check Kaval every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
Alternatives to Kaval
same job, measured the same wayVerify the facts an agent's action depends on before it acts — ALLOW, REVIEW, or BLOCK.
Vet any MCP server before you depend on it. Stamp: PASS, REVIEW, or BLOCK.
Deterministic allow/require_approval/deny verdicts for agent actions, before they happen.
Check whether a real-world fact can be verified before an agent acts on it. Free, no auth.
Check an email before an AI agent sends it: send / ask a human / block. Detects prompt injection.
Validate and verify data from other agents before acting on it. Zero LLM costs.
Explainable ALLOW/WARN/BLOCK risk decisions for AI agent actions before execution, signed as OAA.
The body an agent asks before it acts: decide, approve, and keep a signed record on your machine.
Kaval — questions
Answers built from our own checks of this server.