Attest 1.1.0 released for testing AI agents
Start here
Attest released version 1.1.0, a testing framework for AI agents in Python and TypeScript that checks replies, tool-call sequences and external state changes through pytest or Vitest. It uses a capture, strict replay and gate workflow: strict replay runs the agent's real code against recorded model responses through a local proxy, with no live model calls for matching requests. Checks are split into six layers, from free schema, constraint, trace and content checks to embedding similarity and LLM-as-judge, which depend on a paid or local provider. OpenAI embeddings and OpenAI, Anthropic and Gemini judges are paid, while a local ONNX model, a local ACP agent or an OpenAI-compatible local endpoint are not. Attest advises starting with deterministic checks and adding a model judge only for criteria those checks cannot express.
Generated from reports · updated 14 h ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AIAttest 1.1.0 launches as a Python and TypeScript testing framework for AI agent behavior
Attest 1.1.0 is a testing framework for AI agents in Python and TypeScript that checks replies, tool-call sequences and external state changes through pytest or Vitest. It follows a capture, strict replay and gate workflow, where replay runs the agent's real code against recorded model responses with no live model calls for matching requests. Checks are split into six layers, from free schema, constraint, trace and content checks to embedding similarity and LLM-as-judge, which depend on a paid or local provider.
Heat for this event
- Current, comparable range
- 7
- Peak, comparable range
- 10November 16 03:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.