
Flows
CrewAI · CrewAI
Multi-agent Framework — CrewAI
Evaluates CrewAI's Flows across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Multi-agent Framework eval coverage.
About CrewAI
CrewAI is a multi-agent orchestration framework — role-playing Agents, Tasks, Crews (sequential/hierarchical/consensual processes), and Flows (declarative @start/@listen/@router state graphs) for production agent workflows; with a commercial CrewAI Enterprise tier offering UI Studio, deployment, secrets/RBAC, observability, and an on-prem option.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Flow class declares two @start methods. Operator calls flow.kickoff() expecting one to run. | Multiple @start methods are valid — all fire as parallel entry points. If you want one entry, declare a single @start (or parameterize selection). Verify the actual entry set by inspecting the Flow class. Document which entries exist so kickoff behavior is predictable. | Pass / FailAi Platformhigh |
| 02 | Flow has @listen(method_a) on method_b. method_a returns a value. method_b expects to receive method_a's return. | @listen(target) subscribes method_b to fire when target completes; the return value of target is passed as method_b's first argument. Verify by inspecting the method's signature and the value received. To listen on multiple targets, use the documented or_ / and_ combinators. | Pass / FailAi Platformhigh |
| 03 | @router method returns 'maybe' but no @listen('maybe') is registered. | Default behavior under unmatched @router label is [REQUIRES-VERIFICATION] — the operator should not rely on a silent skip. Either ensure every router label has a listener, add a catch-all listener, or assert at flow-construction time that the router's return set matches the listener set. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Crewai
- Ai Platform
- Flows
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Flows eval for CrewAI CrewAI test?+
Evaluates CrewAI's Flows across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Multi-agent Framework eval coverage.
How is the Flows eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Flows pack for CrewAI CrewAI contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Flows pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.