Evaluate an AI agent across six dimensions: business value, input data, tools and permissions, task evaluation, cost and latency, and operating ownership. Build a representative evaluation set and failure taxonomy before granting wider tools or automatic execution.
A six-dimension scorecard
| Dimension | Question |
|---|---|
| Value | What are frequency, current effort, error cost, and desired outcome? |
| Data | Is data authorized, complete, current, and available? |
| Permissions | What can the agent read or write, and what requires approval? |
| Evaluation | What is correct, how are severe failures graded, and are samples representative? |
| Cost and latency | What are unit cost, peak load, timeout, and fallback? |
| Ownership | Who reviews logs, resolves incidents, updates knowledge, and approves releases? |
Build evals before optimizing prompts
Include normal tasks, edge cases, missing and conflicting information, adversarial inputs, and tool failures. Each case needs an expected result or scoring rule.
Save a baseline and run the same core set whenever models, prompts, retrieval, or tools change.
Start with least privilege
Separate read from write, and draft from publish. Payments, deletion, external messages, and commitments should retain human approval. Stop repeated calls after a failure threshold.
Use release gates
Classify the decision as continue testing, limited release, or pause. Limited release requires acceptable quality, controlled severe failures, viable cost, named ownership, and verified rollback.
Frequently asked questions
Is answer accuracy enough?
No. Include tool behavior, permission violations, severe errors, cost, latency, handoff, and business outcomes.
How large should an eval set be?
There is no fixed size. Cover frequent real tasks and high-risk boundaries, then add production failures.
When may an agent act automatically?
Increase autonomy only after actions are reversible, permissions are minimal, monitoring works, and limited release is stable.
Sources and further reading
- OpenAI Evals guide
Official guidance on continuous evaluation and datasets.
- NIST Generative AI Profile
A reference for identifying and managing generative AI risk.