Open-source reliability platform for detecting, investigating, and explaining AI agent failures.
5.5K
Tessary is an open-source agent reliability platform that monitors every production trace, detects issues that sampled evals miss, and investigates to find the root cause.
Tessary runs cheap, narrow classifiers across every incoming trace. When it detects a recurring issue or drift in behaviour, it groups the affected traces into a finding, validates the issue, and investigates the root cause with your agent codebase.
Run the following command to download and start Tessary:
docker compose -f oci://docker.io/tessaryai/tessary:compose up -d -y
For complete setup instructions, follow the self-hosting guide.