# Tessary classifiers

Tessary's classifiers are the cheap checks that run on every production trace. Sampling misses low-
probability issues, and in a mature agent nearly every remaining issue is low-probability. So
finding them means checking everything, which is only affordable when one check costs almost
nothing.

Each trace is scored by the built-ins: frustration, groundedness, secret_leak, malformed_output,
behavior_drift, duration_drift, cost_drift, and tool_error. When a score crosses its confidence
threshold, a finding is created for further analysis; LLM judgment comes in only at that escalation
step, never by default. Custom classifiers, fine-tuned for your agent's own failure modes, run
alongside them.

Low-probability issues surface as they happen instead of weeks later through user reports, which is
what keeps your critical agents reliable.

## Questions answered under this concept

- [Can I add a custom classifier for my own agent?](https://tessary.ai/answers/tessary-classifiers/can-i-add-a-custom-classifier-for-my-own-agent)
- [How much does it cost to run a classifier on every trace?](https://tessary.ai/answers/tessary-classifiers/how-much-does-it-cost-to-run-a-classifier-on-every-trace)
- [What happens when a Tessary classifier flags a trace?](https://tessary.ai/answers/tessary-classifiers/what-happens-when-a-tessary-classifier-flags-a-trace)
- [What is a Tessary classifier?](https://tessary.ai/answers/tessary-classifiers/what-is-a-tessary-classifier)
- [Why do Tessary's classifiers produce false positives?](https://tessary.ai/answers/tessary-classifiers/why-do-tessary-classifiers-produce-false-positives)
- [Why does Tessary check every trace instead of sampling?](https://tessary.ai/answers/tessary-classifiers/why-does-tessary-check-every-trace-instead-of-sampling)

---

Source: https://tessary.ai/answers/tessary-classifiers
All concepts: https://tessary.ai/answers
From Tessary, agent reliability for AI agents in production: https://tessary.ai