Skip to content

feat(examples): expand CrewAI examples with multi-agent crew and Flow integration#94

Open
joaomdmoura wants to merge 5 commits intoagentcontrol:mainfrom
crewAIInc:feat/crewai-examples-expansion
Open

feat(examples): expand CrewAI examples with multi-agent crew and Flow integration#94
joaomdmoura wants to merge 5 commits intoagentcontrol:mainfrom
crewAIInc:feat/crewai-examples-expansion

Conversation

@joaomdmoura
Copy link
Collaborator

Add 4 new CrewAI example directories demonstrating the full range of Agent Control capabilities:

  • steering_financial_agent: All 3 action types (deny/steer/warn) in a wire-transfer scenario with 5 tested scenarios
  • evaluator_showcase: All 4 built-in evaluators (SQL, LIST, REGEX, JSON) in a data-analyst scenario with 12 tested scenarios
  • secure_research_crew: 3-agent sequential crew (Researcher, Analyst, Writer) with per-role policies and 5 tested scenarios
  • content_publishing_flow: Full CrewAI Flow with @start/@listen/@router, conditional routing (low_risk/high_risk/escalation), embedded crews, client-side steering, and human-in-the-loop with 6 tested scenarios

Update the top-level CrewAI README with comprehensive feature coverage matrix, architecture diagrams, and scenario tables for all 5 examples.

joaomdmoura and others added 4 commits March 11, 2026 01:27
… integration

Add 4 new CrewAI example directories demonstrating the full range of
Agent Control capabilities:

- steering_financial_agent: All 3 action types (deny/steer/warn) in
  a wire-transfer scenario with 5 tested scenarios
- evaluator_showcase: All 4 built-in evaluators (SQL, LIST, REGEX, JSON)
  in a data-analyst scenario with 12 tested scenarios
- secure_research_crew: 3-agent sequential crew (Researcher, Analyst,
  Writer) with per-role policies and 5 tested scenarios
- content_publishing_flow: Full CrewAI Flow with @start/@listen/@router,
  conditional routing (low_risk/high_risk/escalation), embedded crews,
  client-side steering, and human-in-the-loop with 6 tested scenarios

Update the top-level CrewAI README with comprehensive feature coverage
matrix, architecture diagrams, and scenario tables for all 5 examples.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…xamples

Add ?limit=100 to all verify_setup API calls to handle pagination when
many controls exist. Defer LLM creation in evaluator_showcase behind
OPENAI_API_KEY check so examples work without an API key.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…eate layout

Refactor all 4 CrewAI examples (steering_financial_agent, evaluator_showcase,
secure_research_crew, content_publishing_flow) from single-file monoliths to
the standard `crewai create crew/flow` project structure:

- src/<package>/main.py — entry point, scenarios, verify_setup
- src/<package>/crew.py — @crewbase class with YAML config (crew examples)
- src/<package>/config/agents.yaml + tasks.yaml — agent/task definitions
- src/<package>/tools/*.py — tool implementations with @control() wrappers
- pyproject.toml — [tool.crewai] type, [project.scripts] entry points

All examples tested and passing with identical behavior.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…ools dep

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant