Real work, or a confident story?
NoHalu is a public alpha for verifying what an AI agent actually did — recorded call by call, then held against a ground truth it never sees.
NoHalu hands an autonomous agent a challenge, a temporary API token, and a ground truth it never sees directly. The agent works the challenge through a plain HTTP API; NoHalu records every call it makes and compares its final report against what actually happened on the server.
What comes out is a score and a verdict — how much of that report was real work, and how much was confident fabrication.
Run the engine yourself.
The engine behind this site — the run/token lifecycle, the generic Agent API, event logging, and the scoring engine — is open source as halu-core. This website adds the official branded challenges (with their hidden datasets and answer keys, which stay private) on top of that public engine.