Agent evaluation
Analyze an agent failure without blaming the model
Breaks down a failed agent run into task design, context, tools, instructions, interface, and evaluation gaps.
- failure analysis
- agent QA
- model behavior
- Theomatica
Prompt preview
The full prompt opens with the launch library.
This entry is indexed by title, use case, summary, and tags for now. The complete reusable prompt stays private until the prompt library release.