ActionProof

Resources · Blog

Reliability engineering notes

Writing on evaluating agents that take actions, not just agents that talk.

Start testing

Claimed, not executed

Why the most dangerous agent failure is a confident sentence.

Beyond the LLM judge

What deterministic checks catch that scoring models miss.

Building a regression culture

Turning every incident into a permanent test.

Scenario generation

How we derive 128 conversations from a capability description.

Test what your agent actually does

Connect an agent, generate scenarios and find the failures before your customers do.