crewAI

v1.14.7 Security

This release includes 1 security fix for security teams reviewing exposed deployments.

Published 1mo AI Agents & Assistants

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

This release patches 1 known CVE

Topics

agents ai ai-agents aiagentframework llms

Affected surfaces

deps

Summary

AI summary

Updates Bug Fixes, Refactoring, and Performance across a mixed release.

Changes in this release

Type	Severity	Summary	CVE
Security	High	Resolve pip-audit CVEs for aiohttp, docling, and docling-core. Resolve pip-audit CVEs for aiohttp, docling, and docling-core. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Feature
Feature	Low	Add pluggable default backends for memory, knowledge, rag, and flow. Add pluggable default backends for memory, knowledge, rag, and flow. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Surface real finish_reason, sampling params, and response.id on LLM events. Surface real finish_reason, sampling params, and response.id on LLM events. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Add chat API for conversational flows. Add chat API for conversational flows. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Make locking backend overridable. Make locking backend overridable. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Build FlowDefinition from Flow DSL metadata. Build FlowDefinition from Flow DSL metadata. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Add native Snowflake Cortex LLM provider. Add native Snowflake Cortex LLM provider. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Add crew trained agents file support. Add crew trained agents file support. Source: llm_adapter@2026-06-11 Confidence: high	—
Feature	Low	Type DSL triggers as route-aware decorators. Type DSL triggers as route-aware decorators. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Performance	Low	Improve crewai import speed by lazy-loading docling imports. Improve crewai import speed by lazy-loading docling imports. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Bugfix
Bugfix	Medium	Fix checkpoint to rebuild custom BaseLLM as concrete LLM on restore. Fix checkpoint to rebuild custom BaseLLM as concrete LLM on restore. Source: llm_adapter@2026-06-11 Confidence: high	—
Bugfix	Medium	Gate restore on a flag to prevent live snapshots from replaying as resume. Gate restore on a flag to prevent live snapshots from replaying as resume. Source: llm_adapter@2026-06-11 Confidence: high	—
Bugfix	Medium	Scope runtime state per run to bound growth and isolate concurrent runs. Scope runtime state per run to bound growth and isolate concurrent runs. Source: llm_adapter@2026-06-11 Confidence: high	—
Bugfix	Medium	Fix telemetry setup on crewai-login. Fix telemetry setup on crewai-login. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Bugfix	Low	Respect suppress_flow_events for method-execution events. Respect suppress_flow_events for method-execution events. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Bugfix	Low	Restore [project.scripts] in crewai package for uv tool install. Restore [project.scripts] in crewai package for uv tool install. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Bugfix	Low	Fix file input not working reliably. Fix file input not working reliably. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Bugfix	Low	Fix Snowflake Claude incomplete tool result histories. Fix Snowflake Claude incomplete tool result histories. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Refactor
Refactor	Low	Simplify flow condition evaluation to be stateless per event. Simplify flow condition evaluation to be stateless per event. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Refactor	Low	Decouple convo logic from runtime and add a conversational_definition. Decouple convo logic from runtime and add a conversational_definition. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—
Refactor	Low	Split `flow.py` into DSL, definition, and runtime. Split `flow.py` into DSL, definition, and runtime. Source: granite4.1:30b@2026-06-11-audit Confidence: low	—

Full changelog

What's Changed

Features

Add pluggable default backends for memory, knowledge, rag, and flow.
Surface real finish_reason, sampling params, and response.id on LLM events.
Type DSL triggers as route-aware decorators.
Add chat API for conversational flows.
Make locking backend overridable.
Build FlowDefinition from Flow DSL metadata.
Add native Snowflake Cortex LLM provider.
Add crew trained agents file support.

Bug Fixes

Fix checkpoint to rebuild custom BaseLLM as concrete LLM on restore.
Gate restore on a flag to prevent live snapshots from replaying as resume.
Scope runtime state per run to bound growth and isolate concurrent runs.
Fix telemetry setup on crewai-login.
Respect suppress_flow_events for method-execution events.
Restore [project.scripts] in crewai package for uv tool install.
Resolve pip-audit CVEs for aiohttp, docling, and docling-core.
Fix file input not working reliably.
Fix Snowflake Claude incomplete tool result histories.

Documentation

Update changelog and version for v1.14.7.
Update OpenTelemetry collector documentation.
Update NVIDIA Nemotron LLM guide.
Add Databricks integration guide.
Add Snowflake integration guide.

Performance

Improve crewai import speed by lazy-loading docling imports.

Refactoring

Simplify flow condition evaluation to be stateless per event.
Decouple convo logic from runtime and add a conversational_definition.
Split flow.py into DSL, definition, and runtime.

Contributors

@Luzk, @alex-clawd, @devin-ai-integration[bot], @greysonlalonde, @gvieira, @jessemiller, @lorenzejay, @lucasgomide, @mattatcha, @vinibrsl

Security Fixes

Resolved pip-audit CVEs for aiohttp, docling, and docling-core

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track crewAI

Get notified when new releases ship.

About crewAI

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

All releases →