# CrewAI 1.15.18

- Product: CrewAI (https://whatsnew.fyi/product/crewai)
- Vendor: CrewAI
- Date: 2026-08-27
- Version: 1.15.18
- Original notes: https://github.com/crewAIInc/crewAI/releases/tag/1.15.18
- Permalink: https://whatsnew.fyi/product/crewai/releases/1.15.18

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **added** — Promote conversational flows to stable
- **added** — Record a created deployment with the given UUID
- **added** — Enhance conversational flow documentation and APIs
- **added** — Let a declaration name the router's response format
- **added** — Let a chat flow declare its own state shape
- **added** — Accept crew-style LLM config in a conversational declaration
- **added** — Report project creation with the minted ID
- **added** — Record whether a run had inputs, without recording the inputs
- **added** — Backfill project ID from every user-invoked project command
- **fixed** — Preserve tool results when the final answer is empty
- **fixed** — Map default Claude Sonnet 4.6 to its 1M context window
- **fixed** — Raise Anthropic default max_tokens for large tool calls
- **fixed** — Render message content parts as text, not as a Python repr
- **fixed** — Keep message roles when Agent.kickoff gets a conversation
- **fixed** — Skip interception hooks on crewai-internal flows
- **fixed** — Record task failures as failures, not successes
- **fixed** — Emit the flow lifecycle on a suppressed resume
- **fixed** — Open the conversational TUI for a declarative chat flow
- **fixed** — Record crew_memory as a string, not a bool
- **fixed** — Always emit project_id so absent and empty stay distinct

##### What's Changed

###### Features
- Promote conversational flows to stable
- Record a created deployment with the given UUID
- Enhance conversational flow documentation and APIs
- Let a declaration name the router's response format
- Let a chat flow declare its own state shape
- Accept crew-style LLM config in a conversational declaration
- Report project creation with the minted ID
- Record whether a run had inputs, without recording the inputs
- Backfill project ID from every user-invoked project command

###### Bug Fixes
- Preserve tool results when the final answer is empty
- Map default Claude Sonnet 4.6 to its 1M context window
- Raise Anthropic default max_tokens for large tool calls
- Render message content parts as text, not as a Python repr
- Keep message roles when Agent.kickoff gets a conversation
- Skip interception hooks on crewai-internal flows
- Record task failures as failures, not successes
- Emit the flow lifecycle on a suppressed resume
- Open the conversational TUI for a declarative chat flow
- Record crew_memory as a string, not a bool
- Always emit project_id so absent and empty stay distinct

###### Documentation
- Clarify Arize Phoenix observability docs

##### Contributors

@Vidit-Ostwal, @arizedatngo, @joaomdmoura, @lorenzejay, @lucasgomide