1.15.2

feder-cr/AIHawk1.15.2Jul 8, 2026by lorenzejay

AI Summary

Enhanced the CrewAI agent framework with dynamic LLM model loading, templated flow actions, and improved message handling in the AgentExecutor.

Key Highlights

  • Pull latest LLM models dynamically in the crew wizard
  • Support for inline skill definitions and templated Flow action inputs
  • Added generated Flow Definition authoring skill
  • Implemented stream frame protocol for flows and repository agents

New Features

  • Dynamic LLM model loading
  • Inline skill definitions
  • Templated Flow action inputs
  • Stream frame protocol for flows

Full Release Notes

## What's Changed

### Features
- Pull latest LLM models dynamically in the crew wizard.
- Support inline skill definitions.
- Add generated Flow Definition authoring skill.
- Support templated Flow action inputs.
- Add text helper for flow CEL prompts.
- Add text helper to flow skill example.
- Implement message setup and feedback handling in AgentExecutor.
- Add repository agents to flow definitions.
- Define stream frame protocol for flows.
- Type tool and app in CrewDefinition.
- Repoint template commands to crewAIInc-fde org.

### Bug Fixes
- Key model-catalog cache by exact API key, shorten TTL, and skip Ollama.
- Unify `crewai run` flow input resolution and prompt from the state schema.
- Resolve pip-audit failures for onnx 1.22.0 and nltk PYSEC-2026-597.
- Ensure we are writing version for flows.
- Include aiobotocore in the bedrock extra.
- Reject self-listening flow methods.
- Cut docs version nav from Edge so new pages aren't dropped.

### Documentation
- Update language from Rules to Policies to match new dashboard changes.
- Document flow agent options.
- Add streaming docs to the navigation.
- Document Cost Limit rule type in Agent Control Plane.
- Drop CREWAI_LOG_FORMAT references from Datadog guide.

## Contributors

@akaKuruma, @danielfsbarreto, @github-code-quality[bot], @joaomdmoura, @lorenzejay, @lucasgomide, @manisrinivasan2k1, @renatonitta, @vinibrsl