◆ AI Agent Frameworks

CrewAI

1.15.18 open-source

CrewAI is a framework for building multi-agent AI systems where agents collaborate to complete complex tasks autonomously.

source ↗ rss ↗ last repo push Aug 19, 2026

crewai create <resource>
from crewai_tools import URLReadTool

url_tool = URLReadTool()

agent = Agent(
    role='Web Researcher',
    goal='Gather information from public URLs',
    backstory='An expert at reading web content',
    tools=[url_tool]
)
from crewai import on

@on('after_llm_call')
def inspect_llm_output(context):
    print(context)
from crewai.flow.flow import Flow

flow = Flow()
response = flow.chat("What is the status of my order #1234?")
result = my_crew.kickoff(restore_from_state_id="<state_id>")
from crewai import LLM

# No api_key provided — CrewAI falls back to DefaultAzureCredential automatically
llm = LLM(
    model="azure/<your-deployment>",
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    api_version="2024-02-01",
)
result = crew.kickoff(from_checkpoint=<checkpoint_id>)
crewai checkpoint list
crewai checkpoint info <checkpoint-id>
from crewai.events import LLMCallCompletedEvent
from crewai.event_bus import event_bus

@event_bus.on(LLMCallCompletedEvent)
def on_llm_done(event: LLMCallCompletedEvent):
    print(event.token_usage)
from crewai import LLM

llm = LLM(
    model="ollama/llama3",
    base_url="http://localhost:11434"
)
crewai logout
flow = MyFlow()
structure = flow.flow_structure()
print(structure)
from pydantic import BaseModel
from crewai import Agent, Task, Crew

class ThreatReport(BaseModel):
    severity: str
    affected_systems: list[str]
    recommendation: str

analyst = Agent(role="Threat Analyst", goal="Analyze security incidents", backstory="...", verbose=False)
task = Task(
    description="Analyze the attached incident log and produce a threat report.",
    expected_output="A structured threat report",
    agent=analyst,
    response_format=ThreatReport,
)
crew = Crew(agents=[analyst], tasks=[task])
result = crew.kickoff()
report: ThreatReport = result.pydantic
crewai bump --no-commit
import asyncio
from crewai import Crew, Agent, Task

agent = Agent(role="Analyst", goal="Analyze data", backstory="Expert analyst")
task = Task(description="Summarize the report", agent=agent, expected_output="Summary")
crew = Crew(agents=[agent], tasks=[task])

async def main():
    result = await crew.kickoff_async()
    print(result)

asyncio.run(main())
import asyncio
from crewai.flow.flow import Flow, start

class MyFlow(Flow):
    @start()
    async def run_step(self):
        return "done"

async def main():
    flow = MyFlow()
    result = await flow.kickoff_async()
    print(result)

asyncio.run(main())
result = crew.kickoff()
for task_output in result.tasks_output:
    print(task_output.messages)
embedder_config = {
    "provider": "openai",
    "config": {
        "model": "text-embedding-3-small",
        "batch_size": 50
    }
}
crewai config reset
crewai config reset
crewai enterprise configure
crewai config
from crewai import Task

task = Task(
    description="Summarize the latest threat intelligence report.",
    expected_output="A structured summary of key findings.",
    markdown=True
)
from crewai import Agent

analyst = Agent(
    role="Threat Analyst",
    goal="Identify emerging threats from recent feeds.",
    backstory="Expert in cyber threat intelligence.",
    reasoning=True,
    inject_date=True
)
from crewai.tools import tool

@tool(result_as_answer=True)
def lookup_cve(cve_id: str) -> str:
    """Fetch CVE details from internal database."""
    return fetch_cve_record(cve_id)
from crewai import Agent

researcher = Agent(
    role="Research Analyst",
    goal="Find the latest CVEs for Apache HTTP Server",
    backstory="You are an expert in vulnerability research.",
    llm="gpt-4o"
)

result = researcher.kickoff()
print(result)
from crewai_tools import QdrantVectorSearchTool

tool = QdrantVectorSearchTool(
    collection_name="security-advisories",
    url="http://localhost:6333",
    api_key="<your-qdrant-api-key>"
)
agent = Agent(role="Threat Analyst", tools=[tool], ...)
from crewai.flow.persistence import persist, FlowPersistence

class MySecurityFlow(Flow):
    @persist
    def analyze_targets(self):
        # state is automatically saved after this method completes
        ...
@crew
def my_crew(self) -> Crew:
    return Crew(
        agents=self.agents,
        tasks=self.tasks,
        before_kickoff=self.before_kickoff,
        after_kickoff=self.after_kickoff,
    )
from crewai import BaseTool

class MyScanner(BaseTool):
    name: str = "Port Scanner"
    description: str = "Scans open ports on a given host."

    def _run(self, host: str) -> str:
        # tool logic here
        return f"Scanning {host}"
crewai flow add-crew
crewai flow plot
crewai create flow
crewai tool create <tool>
from crewai import Agent, LLM

llm = LLM(model="gpt-4o")
agent = Agent(
    role="Security Analyst",
    goal="Identify vulnerabilities in the provided code.",
    backstory="Expert in application security.",
    llm=llm
)
from crewai import Agent

agent = Agent(
    role="Analyst",
    goal="Analyze the dataset",
    backstory="You are a data expert.",
    llm="o1-preview",
    use_system_prompt=False,
    use_stop_words=False
)
from crewai import Agent

agent = Agent(
    role="Researcher",
    goal="Find key insights",
    backstory="You are a senior researcher.",
    max_rpm=10,
    max_iter=5
)
crewai install
crewai deploy
result = crew.kickoff()
for task_output in result.tasks_output:
    print(task_output.name, task_output.expected_output)
from crewai import Crew, Agent, Task, Process
from langchain_openai import ChatOpenAI

planner_llm = ChatOpenAI(model="gpt-4o-mini")

crew = Crew(
    agents=[...],
    tasks=[...],
    process=Process.sequential,
    planning=True,
    planning_llm=planner_llm
)
crew.kickoff()
from crewai_tools import DallETool
from crewai import Agent

image_agent = Agent(
    role="Image Creator",
    goal="Generate visuals from descriptions",
    backstory="You create images for marketing campaigns.",
    tools=[DallETool()]
)
from crewai_tools import NL2SQLTool
from crewai import Agent

db_agent = Agent(
    role="Data Analyst",
    goal="Answer business questions from the database",
    backstory="You query databases using plain English.",
    tools=[NL2SQLTool(db_uri="mysql+pymysql://user:pass@host/dbname")]
)
from crewai import Crew

crew = Crew(
    agents=[...],
    tasks=[...],
    planning=True
)
result = crew.kickoff()
crewai replay <task_id>
crewai reset-memory
crewai train -n 5
results = my_crew.kickoff_for_each(inputs=[{"target": "host1"}, {"target": "host2"}, {"target": "host3"}])
import asyncio

async def main():
    result = await my_crew.kickoff_async(inputs={"target": "host1"})
    print(result)

asyncio.run(main())
result = my_crew.kickoff(inputs={"target": "host1"})
print(result.usage_metrics)
crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    memory=True
)
from crewai_tools import tool

@tool
def my_tool(query: str) -> str:
    ...

my_tool.cache_function = lambda args, result: 'volatile' not in args['query']
crewai create
from crewai import Agent, Task, Crew

agent = Agent(**{
    "role": "Researcher",
    "goal": "Find the latest AI news",
    "backstory": "You are an expert at finding information."
})

task = Task(**{
    "description": "Search for the top 5 AI news stories today",
    "agent": agent
})
result = crew.kickoff(inputs={'domain': 'example.com'})
print(crew.usage_metrics)
agent = Agent(
    role='Analyst',
    llm=my_llm,
    function_calling_llm=my_function_llm,
    tools=[my_tool]
)
def my_callback(step):
    print(step)

agent = Agent(..., step_callback=my_callback)
crew = Crew(agents=[agent], ..., step_callback=my_callback)
from crewai import Agent, Task, Crew, Process

researcher = Agent(
    role='Researcher',
    goal='Find key facts about quantum computing',
    tools=[search_tool]
)

writer = Agent(
    role='Writer',
    goal='Summarize research into a short briefing',
    tools=[]
)

research_task = Task(description='Research recent quantum computing breakthroughs', tools=[search_tool])
write_task    = Task(description='Write a 200-word executive summary', tools=[])

crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    process=Process.sequential
)

result = crew.kickoff()
print(result)

Summary

CrewAI is an open-source (MIT-licensed) Python framework for building multi-agent automation, letting developers define agents with roles that collaborate autonomously through "Crews" or follow explicit, event-driven logic through "Flows." It's a library imported into application code rather than a standalone service, aimed at developers building LLM-driven automation who need more structure than a single agent loop provides — flows support conditional branching, checkpointed state, and CLI-driven kickoff and resumption. Its own README frames it as a production-ready alternative to piecing together agent orchestration by hand, with pluggable backends for memory, knowledge, and RAG rather than a fixed stack. Development is active: 392 contributors, over 2,000 commits in the past year, and a release 16 days ago.

CrewAI is a framework for building multi-agent AI systems where agents collaborate to complete complex tasks autonomously.

What CrewAI answers

Can I recover a run that failed partway through, or does it start over from the beginning?

crew and flow state can be checkpointed and persisted, and a run resumes from a saved state identifier instead of restarting

What happens to my code if I upgrade?

removed accessors and renamed tools break existing code that references the old names, so upgrades need a pass over agent and tool references

Can I swap out the memory or retrieval backend, or am I locked into one stack?

memory, knowledge, RAG, and the flow locking backend are pluggable defaults, each replaceable rather than fixed

Do I have to write code to define a flow, or can it come from configuration?

flows and crews can be declared in config and loaded rather than authored entirely in code, and kicked off from the command line

Can I see what a run actually cost in tokens?

each LLM call emits usage and completion data, giving per-call cost and rate-limit visibility rather than an aggregate estimate

Is this still an experimental project or something I can build against?

most capabilities ship as regular releases, though newer additions like the skills repository sit behind an experimental gate until stabilized

all 29 features, with the evidence for each →

Features

29 capabilities across 6 areas · 1 backed by code, an API document or a real run

Built from everything we hold on CrewAI — every release we have summarised, its product documentation and how that documentation has changed, its README, its command-line surface and API, and runs we performed ourselves. Dates are when we first saw a capability, not when the vendor introduced it.

Capability area

Flow Authoring and Orchestration 8 capabilities CrewAI Flows let users define, configure, and run multi-step AI workflows declaratively, without writing imperative code for every transition. Conditional routing, iteration, composite actions, and human feedback steps are all expressible within a single flow definition.

Flow composite actions shipped Flow definitions support crew steps, single-agent calls, script blocks, and looping iteration with optional filtering as first-class action types. 10 releases · first seen Jun 2026

release 10 total

Adds templated Flow action inputs, enabling dynamic parameterization of Flow actions at runtime. 1.15.2 · Jul 2026 · source · release history
Adds repository agents to Flow definitions, enabling agent-backed steps directly in Flow authoring. 1.15.2 · Jul 2026 · source · release history
Adds single-agent action support in Flow definitions, allowing a lone agent call to be wired directly into a flow without a full crew. 1.15.0 · Jun 2026 · source · release history
Adds each composite action to FlowDefinition for iterating over collections within a flow step. 1.15.0 · Jun 2026 · source · release history
Adds optional if expression to each.do steps, enabling conditional execution inside iteration blocks. 1.15.0 · Jun 2026 · source · release history
Adds crew actions to FlowDefinition, enabling inline crew execution as a named step in a declarative flow. 1.15.0 · Jun 2026 · source · release history

AI coding agent authoring assistance shipped A built-in authoring skill guides AI coding agents when scaffolding or editing Flow definitions. 7 releases · 2 other sources · first seen Jun 2026

release 7 total

Adds progressive disclosure for skills, surfacing contextually relevant skill guidance incrementally rather than all at once. 1.15.9 · Jul 2026 · source · release history
Emits skill usage events at runtime, enabling observability pipelines to trace which skills agents invoke during execution. 1.15.7 · Jul 2026 · source · release history
Emits skill usage events at runtime, enabling observability pipelines to track which skills agents invoke during execution. 1.15.7a1 · Jul 2026 · source · release history
Supports authenticated downloads from the skill registry, enabling access to private or restricted skills. 1.15.5 · Jul 2026 · source · release history
Promotes the Skills Repository out of experimental status, making structured coding-agent instructions (for Claude Code, Cursor, Windsurf, etc.) a stable, supported capability. 1.15.4 · Jul 2026 · source · release history
Adds a generated Flow Definition authoring skill to guide AI coding agents when scaffolding Flows. 1.15.2 · Jul 2026 · source · release history

example

Install the stable CrewAI Skills into Cursor, Codex, Windsurf, or any other coding agent via the skills.sh registry. $ npx skills add crewaiinc/skills 1.15.4 · Jul 2026 · source
Install the now-stable CrewAI Skills into Claude Code so it automatically applies CrewAI best practices when you scaffold Flows, configure Crews, or design agents. $ /plugin marketplace add crewAIInc/skills /plugin install crewai-skills@crewai-plugins /reload-plugins 1.15.4 · Jul 2026 · source

Declarative Flow definitions shipped Users can author and run complete Flows from JSON or DSL config files instead of writing Python code. 7 releases · first seen Jun 2026

release 7 total

Adds unified declarative Flow loading, letting flows be defined and loaded from configuration rather than pure Python code. 1.15.0 · Jun 2026 · source · release history
Enables declarative references to be resolved across Flow and Crew boundaries, allowing shared state and configuration to be referenced from either context. 1.14.8a5 · Jun 2026 · source · release history
Adds inline crew definition loading so a crew can be fully specified inside a FlowDefinition without a separate crew config file. 1.15.0 · Jun 2026 · source · release history
Adds unified declarative Flow loading, enabling Flows to be defined and loaded via configuration rather than only imperative Python. 1.14.8a3 · Jun 2026 · source · release history
Wires config and persistence settings from FlowDefinition into the runtime, closing the gap between declarative definitions and live execution. 1.14.8a · Jun 2026 · source · release history
Introduces JSON-first crew projects, enabling crew creation, configuration, and execution entirely through JSON rather than Python. 1.14.8a · Jun 2026 · source · release history

Flow DSL conditional routing shipped Flows support conditional branching using route-aware decorators and CEL expressions, including using a router as the entry point. 4 releases · 2 other sources · first seen Jun 2026

release

Adds text helpers for Flow CEL prompts and skill examples, streamlining expression authoring in Flows. 1.15.2 · Jul 2026 · source · release history
Enables @router() to be used as the start method of a Flow, allowing conditional branching logic at Flow entry points. 1.14.8a3 · Jun 2026 · source · release history
Validates Flow CEL (Common Expression Language) expressions at definition load time, catching misconfigured routing conditions before runtime. 1.14.8a2 · Jun 2026 · source · release history
Introduces route-aware decorators for typing DSL triggers in Flows, enabling cleaner conditional routing definitions. 1.14.7 · Jun 2026 · source · release history

example

Define a Flow whose entry point is a router for immediate conditional branching at startup. $ from crewai.flow.flow import Flow, router, listen, start class TriageFlow(Flow): @router() @start() def triage(self): if self.state.priority == 'high': return 'urgent' return 'normal'… 1.14.8a3 · Jun 2026 · source
Route a flow's next step based on a condition using the new route-aware DSL decorator syntax. $ from crewai.flow.flow import Flow, listen, router, start class TriageFlow(Flow): @start() def ingest(self): self.state.severity = classify_alert() @router(ingest) def route_by_severity(self): if… 1.14.7 · Jun 2026 · source

Conversational Flows shipped Flows support multi-turn conversational patterns, including a chat API and terminal UI for interactive dialogue. 4 releases · first seen Jun 2026

release

Tracks conversational flow turn usage in telemetry for visibility into multi-turn conversation costs. 1.15.0 · Jun 2026 · source · release history
Supports conversational flows in the CLI TUI, surfacing turn-by-turn flow progress in the terminal UI. 1.15.0 · Jun 2026 · source · release history
Supports conversational flows in the CLI text-based UI (TUI), enabling interactive multi-turn flow execution from the terminal. 1.14.8a4 · Jun 2026 · source · release history
Adds a chat API for building conversational flows, enabling back-and-forth interaction patterns within CrewAI Flows. 1.14.7 · Jun 2026 · source · release history

Human-in-the-loop feedback in Flows shipped Human feedback steps can be embedded directly in a flow definition without writing custom callback code. 2 releases · first seen Jun 2026

release

Implements message setup and feedback handling in AgentExecutor for richer agent interaction loops. 1.15.2 · Jul 2026 · source · release history
Enables human feedback to be driven directly from a flow definition, removing the need for custom Python callbacks. 1.14.8a · Jun 2026 · source · release history

Direct flow state access shipped Flow methods access flow state directly, removing an indirection layer and simplifying state handling. 2 releases · first seen Jun 2026

release

The StateProxy has been removed from flow state access 1.15.0 · Jun 2026 · source · release history
Removes StateProxy from flow state access, providing direct state access in flow methods (breaking change). 1.15.0 · Jun 2026 · source · release history

DMN (Decision Model and Notation) support shipped Crews can run in DMN mode, letting users model structured decision logic alongside standard agent workflows. 2 releases · first seen Jun 2026

release

Implements DMN (Decision Model and Notation) mode support in crew creation and execution for structured decision logic. 1.15.0 · Jun 2026 · source · release history
Implements DMN (Decision Model and Notation) mode support in crew creation and execution. 1.14.8a · Jun 2026 · source · release history

Crew and Agent Configuration 6 capabilities Crews and agents can be fully defined through structured configuration rather than code, covering agent skills, tool schemas, training state, and artifact types. This makes projects portable and easier to manage across environments.

JSON-first crew projects shipped Crews can be created, configured, and packaged entirely through JSON definitions without writing Python. 3 releases · first seen Jun 2026

release

Pulls latest LLM models dynamically in the crew wizard, keeping model selection current without manual updates. 1.15.2 · Jul 2026 · source · release history
Introduces JSON-first crew projects, enabling crew creation, configuration, and execution entirely through JSON rather than Python. 1.14.8a · Jun 2026 · source · release history
Supports ZIP deployment fallback and JSON crew project environment runs, enabling portable crew packaging and deployment. 1.14.8a · Jun 2026 · source · release history

Typed tool output schemas shipped Tools can declare typed output schemas so that results are structured and validated when returned to an agent. 2 releases · first seen Jun 2026

release

Surfaces tool failures as actual failures instead of silently reporting them as success, enabling agents to react correctly to tool errors. 1.15.9 · Jul 2026 · source · release history
Adds typed output schemas for CrewAI tools, enabling structured, validated outputs from tool calls. 1.14.8a3 · Jun 2026 · source · release history

CrewDefinition artifact types shipped CrewDefinition supports tool and app artifact type declarations, broadening what kinds of crew artifacts can be specified. 2 releases · first seen Jun 2026

release

Types tool and app fields in CrewDefinition, providing stronger schema validation for crew configs. 1.15.2 · Jul 2026 · source · release history
Adds tool and app type options to CrewDefinition, expanding the kinds of crew artifacts that can be declared. 1.15.2a1 · Jun 2026 · source · release history

Inline skill definitions shipped Skills can be declared inline within agent or crew configurations without needing separate skill files. 2 releases · first seen Jun 2026

release

Supports inline skill definitions, eliminating the need for separate skill files in crew/flow configurations. 1.15.2 · Jul 2026 · source · release history
Supports inline skill definitions, enabling skills to be declared directly within agent or crew configurations without separate files. 1.15.2a1 · Jun 2026 · source · release history

Memory management shipped Memory can be reset on demand, with reliable handling for JSON-configured crews. 1 release · first seen Jun 2026

release

Enhances memory reset functionality with improved JSON crew handling. 1.14.8a · Jun 2026 · source · release history

Agent training state persistence shipped Agent training state can be saved and reloaded from a file, preserving learned behaviour across sessions. 1 release · first seen Jun 2026

release

Adds crew trained agents file support for persisting and reloading agent training state. 1.14.7 · Jun 2026 · source · release history

Developer Tooling and CLI 5 capabilities The CrewAI CLI covers the full development lifecycle from scaffolding new projects to deploying and running them. A terminal UI surfaces execution progress, and cold-start optimisations keep large deployments responsive.

Flow CLI execution verified Flows and crews can be started directly from the command line, including flows defined in config files. 5 releases · 4 other sources · first seen Jun 2026

release

Consolidates crewai run and crewai flow kickoff into a single command entry point (breaking change for scripts that invoked them separately). 1.15.0 · Jun 2026 · source · release history
Adds declarative Flow CLI support so flows defined in config files can be kicked off directly from the command line. 1.15.0 · Jun 2026 · source · release history
Adds declarative Flow CLI support, letting practitioners launch declaratively defined Flows from the command line. 1.14.8a3 · Jun 2026 · source · release history
Consolidates crewai run and crewai flow kickoff into a single command, simplifying the CLI surface. 1.14.8a3 · Jun 2026 · source · release history
Adds crewai run --definition experimental CLI flag to run flows directly from a flow definition file without writing Python code. 1.14.8a · Jun 2026 · source · release history

command line

--n-iterations — Number of iterations to Test the crew 1.15.15 · Aug 2026 · command-line history
--task-id — Replay the crew from this task ID, including all subsequent tasks. 1.15.15 · Aug 2026 · command-line history
--skip-provider — Skip provider validation 1.15.15 · Aug 2026 · command-line history

example

Run a flow defined entirely in a JSON/definition file — no Python code required — to execute a crew-based automation. $ crewai run --definition 1.14.8a · Jun 2026 · source

CLI project scaffolding shipped The CLI scaffolds new CrewAI projects with structured definitions and initialises a Git repository automatically. 4 releases · 1 other source · first seen Jun 2026

release

Unifies project scaffolding under crewai create <resource>, replacing separate scaffold commands. 1.15.12 · Aug 2026 · source · release history
Requires explicit CrewAI project definitions, enforcing structured crew configuration at project creation. 1.15.1 · Jun 2026 · source · release history
Automatically initializes a Git repository when generating a new CrewAI project via the CLI. 1.15.1 · Jun 2026 · source · release history
Requires explicit CrewAI project definitions, enforcing stricter project structure as a breaking change for existing setups. 1.15.1a1 · Jun 2026 · source · release history

example

Scaffold a new CrewAI resource (crew, flow, etc.) using the unified create command. $ crewai create <resource> 1.15.12 · Aug 2026 · source

CLI terminal UI (TUI) shipped A terminal UI shows crew and flow execution progress, conversational output, and interactive controls without leaving the terminal. 4 releases · first seen Jun 2026

release

Supports running declarative Flows on the TUI with headless terminal fallback. 1.15.3 · Jul 2026 · source · release history
Supports running declarative flows on the TUI with a headless terminal fallback. 1.15.3a1 · Jul 2026 · source · release history
Adds telemetry tracking for TUI button interactions, enabling usage analytics on crew terminal UI actions. 1.15.1a1 · Jun 2026 · source · release history
Supports conversational flows in the CLI TUI, surfacing turn-by-turn flow progress in the terminal UI. 1.15.0 · Jun 2026 · source · release history

CLI deployment integration shipped After a successful deploy command, the deployment page opens automatically in the browser. 2 releases · first seen Jun 2026

release

Opens the deployment page automatically in the browser after a successful crewai deploy command. 1.15.1 · Jun 2026 · source · release history
Automatically opens the deployment page in the browser after a successful crewai deploy CLI command. 1.15.1a1 · Jun 2026 · source · release history

Cold-start performance optimisation shipped Lazy-loading of a heavy dependency reduces import time and cold-start latency in large deployments. 1 release · first seen Jun 2026

release

Improves import speed via lazy-loading of docling — meaningfully reducing cold-start latency in large deployments. 1.14.7 · Jun 2026 · source · release history

Observability and Execution Control 4 capabilities CrewAI exposes hooks, streaming events, and LLM-level telemetry so teams can monitor, audit, and gate what agents do at runtime. Interception points allow logging or blocking at execution boundaries without modifying core logic.

Flow observability shipped Flow execution surfaces nested crew progress, real LLM finish reasons, sampling parameters, and response IDs for detailed monitoring. 7 releases · 1 other source · first seen Jun 2026

release 7 total

Reports flow outcome, duration, and human-in-the-loop signals as part of flow telemetry. 1.15.15 · Aug 2026 · source · release history
Tracks interception-hook dispatches in telemetry for improved observability of agent execution. 1.15.11 · Aug 2026 · source · release history
Emits a new FlowFailedEvent when a flow execution fails, allowing event-driven error handling and observability hooks in Flows. 1.15.9 · Jul 2026 · source · release history
Adds step interception points and a generic hook dispatcher, enabling fine-grained execution-boundary callbacks via @on decorators. 1.15.3 · Jul 2026 · source · release history
Adds step interception points and a generic hook dispatcher, enabling @on-decorated callbacks at execution boundaries for fine-grained observability and control. 1.15.3a1 · Jul 2026 · source · release history
Keeps flow method progress visible for nested Crews, improving observability when Crews are embedded inside Flows. 1.14.8a3 · Jun 2026 · source · release history

example

Intercept execution boundaries in a crew or flow to log, audit, or gate agent activity at specific lifecycle points. $ from crewai import on @on("before_llm_call") def audit_llm_input(context): print("LLM about to be called with:", context) @on("after_llm_call") def log_llm_output(context): print("LLM responded… 1.15.3 · Jul 2026 · source

Execution hook interception shipped Decorated callbacks fire at execution boundaries, enabling logging, auditing, or gating of agent activity without modifying core code. 2 releases · first seen Jul 2026

release

Adds step interception points and a generic hook dispatcher, enabling fine-grained execution-boundary callbacks via @on decorators. 1.15.3 · Jul 2026 · source · release history
Adds step interception points and a generic hook dispatcher, enabling @on-decorated callbacks at execution boundaries for fine-grained observability and control. 1.15.3a1 · Jul 2026 · source · release history

Flow streaming shipped A structured stream frame protocol defines how Flow execution events are formatted and delivered to consumers in real time. 2 releases · first seen Jun 2026

release

Defines a stream frame protocol for Flows, enabling structured streaming output from Flow executions. 1.15.2 · Jul 2026 · source · release history
Introduces a stream frame protocol for Flows, defining a structured message format for streaming Flow execution events. 1.15.2a1 · Jun 2026 · source · release history

WaitTool for long-running background jobs shipped A WaitTool lets agents pause execution until a long-running background job finishes before continuing. 1 release · first seen Jul 2026

release

Adds WaitTool to pause agent execution while waiting on long-running background jobs. 1.15.8 · Jul 2026 · source · release history

Integrations and Data Access 3 capabilities CrewAI provides built-in tools and provider integrations that give agents access to external data sources and LLM backends. These cover database querying, web content retrieval, and enterprise LLM platforms.

Snowflake Cortex LLM provider shipped Snowflake Cortex is available as a native LLM provider for agents and crews. 1 release · 1 other source · first seen Jun 2026

release

Adds native Snowflake Cortex as a supported LLM provider. 1.14.7 · Jun 2026 · source · release history

example

Use Snowflake Cortex as the LLM provider for an agent that queries internal data warehouses. $ from crewai import Agent, LLM analyst = Agent( role="Data Analyst", goal="Summarize query results from Snowflake", backstory="Expert in SQL and data warehousing.",… 1.14.7 · Jun 2026 · source

IBM Db2 search tool shipped A built-in IBM Db2 tool lets agents query Db2 databases directly within a workflow. 1 release · first seen Aug 2026

release

Adds IBM Db2 search tool, enabling agents to query Db2 databases as a built-in tool. 1.15.11 · Aug 2026 · source · release history

URL reading tool shipped A URL reading tool lets agents fetch and read content from arbitrary web URLs as part of a workflow. 1 release · first seen Aug 2026

release

Adds URLReadTool for reading arbitrary URLs inside agent workflows. 1.15.12 · Aug 2026 · source · release history

Platform Architecture and Extensibility 3 capabilities Core backends for memory, knowledge, storage, and locking are swappable, letting operators plug in custom engines without subclassing internals. Multi-org support and enterprise project linking connect open-source usage to managed platform accounts.

Pluggable backends shipped Default backends for memory, knowledge, RAG, flow storage, and locking can be replaced with custom implementations without changing core code. 3 releases · first seen Jun 2026

release

Improves custom OpenAI URL handling for self-hosted or proxy LLM endpoints. 1.15.3a1 · Jul 2026 · source · release history
Makes the locking backend overridable for concurrent-run isolation and custom synchronization strategies. 1.14.7 · Jun 2026 · source · release history
Introduces pluggable default backends for memory, knowledge, RAG, and flow — allowing custom storage and retrieval engines without subclassing core internals. 1.14.7 · Jun 2026 · source · release history

Multi-org platform support shipped The platform API client accepts an organisation ID, enabling a single deployment to serve multiple organisations. 2 releases · first seen Jul 2026

release

Adds organization_id parameter to the PlusAPI client for multi-org deployments. 1.15.3 · Jul 2026 · source · release history
Adds an organization ID parameter to the PlusAPI client for multi-org deployments. 1.15.3a1 · Jul 2026 · source · release history

Enterprise project linking shipped A project ID parameter links an open-source CrewAI installation to an enterprise platform account. 1 release · first seen Aug 2026

release

Adds project_id support to link open-source CrewAI usage to an enterprise account. 1.15.11 · Aug 2026 · source · release history

Capability

Evidence

Lines in monospace are the tool's own words — help text parsed from its source, or an endpoint from its API document. Everything else is our summary of a dated release or documentation change, linked back to the source it came from.

Command line

composed command

option surface across analysed releases

options at that release added, per release removed, per release

option	found in	since	description

Values are placeholders taken from each option’s declared default. Nothing is executed here — the output shown is a recording of a run that already happened.

Release history

1.15.18 Aug 27, 2026 · issue 009

CrewAI 1.15.18 promotes conversational flows to stable, adds declarative state/router config, and crew-style LLM config in chat declarations.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.18 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.18
```
- ›Conversational flows are now stable (promoted out of experimental/beta status).
- ›A declarative chat flow can now define its own state shape directly in the flow declaration.
- ›A router's response format can now be named in the flow declaration itself.
- ›Crew-style LLM config is now accepted inside a conversational flow declaration, unifying configuration patterns.
- ›Deployment creation now records the deployment with a given UUID, and project creation is reported with the minted project ID.
+2 moreshow less
- ›Run telemetry now records whether a run had inputs, without recording the input values themselves.
- ›Project ID is backfilled from every user-invoked project command.
1.15.17 Aug 20, 2026 · issue 002

CrewAI 1.15.17 adds declarative conversational flow support and AMP slug propagation for tool resolution.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.17 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.17
```
- ›Enables declarations to drive conversational mode, making conversational flows opt-in via declarative configuration rather than imperative code.
- ›Synthesizes built-in conversational methods for declarations, allowing conversational agent behaviors to be expressed in YAML/declarative form.
- ›Handles oversized single messages during chunking, enabling processing of payloads that previously exceeded per-message size limits.
1.15.16 Aug 14, 2026 · issue -005

CrewAI 1.15.16 adds execution context management with UUID support and richer observability for flows and deployments.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.16 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.16
```
- ›Adds execution context management with UUID support for tracking individual execution runs.
- ›Records the type of exception that ended a flow, enabling richer post-mortem observability.
- ›Records when a trace batch is shared with AMP, improving telemetry auditability.
- ›Counts deployments from any origin and records where they started, broadening deployment provenance tracking.
1.15.16 Aug 14, 2026 · issue 001

CrewAI 1.15.16 adds execution context management with UUID support, flow exception recording, and expanded deployment origin tracking.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.16 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.16
```
- ›Introduces execution context management with UUID support for tracing individual flow runs.
- ›Records the type of exception that ended a flow, enabling richer post-mortem observability in AMP traces.
- ›Tracks when a trace batch is shared with AMP, improving observability auditability.
- ›Counts deployments from any origin and records where each deployment started, expanding deployment analytics across all launch surfaces.
1.15.15 Aug 12, 2026 · issue -007

CrewAI 1.15.15 adds flow outcome, duration, and human-in-the-loop signal reporting to flow telemetry.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.15 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.15
```
- ›Reports flow outcome, duration, and human-in-the-loop signals as part of flow telemetry.
1.15.15 Aug 12, 2026 · issue 001

CrewAI 1.15.15 adds flow outcome, duration, and human-in-the-loop signal reporting to flow telemetry.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.15 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.15
```
- ›Flow execution now reports outcome, duration, and human-in-the-loop signals as part of flow telemetry.
1.15.14 Aug 8, 2026 · issue 001

CrewAI 1.15.14 splits runtime context from the coding agent and introduces project ID support.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.14 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.14
```
- ›Separates runtime context from the coding agent, allowing project ID to be tracked independently per execution.
1.15.14 Aug 8, 2026 · issue -011

CrewAI 1.15.14 splits runtime context from the coding agent and introduces project ID support.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.14 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.14
```
- ›Separates runtime context from the coding agent, allowing project ID to be tracked independently per execution.
1.15.12 Aug 5, 2026 · issue -014

CrewAI 1.15.12 adds URLReadTool, app metadata on platform action tools, and unifies scaffolding under crewai create
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.12 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.12
```
└──▷ TRY IT
Scaffold a new CrewAI resource using the unified create command instead of legacy subcommands.
```
$ crewai create <resource>
```
- ›Unifies scaffolding commands under the crewai create <resource> subcommand.
- ›Adds URLReadTool for reading arbitrary URLs.
- ›Adds app metadata to platform action tools.
1.15.12 Aug 5, 2026 · issue 001

CrewAI 1.15.12 adds URLReadTool, unifies project scaffolding under crewai create <resource>, and exposes app metadata on platform action tools.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.12 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.12
```
└──▷ USE IT
Give an agent the ability to retrieve live web content by attaching URLReadTool to its tool list.
python
```
from crewai_tools import URLReadTool

url_tool = URLReadTool()

agent = Agent(
    role='Web Researcher',
    goal='Gather information from public URLs',
    backstory='An expert at reading web content',
    tools=[url_tool]
)
```
- ›Unifies project scaffolding under the crewai create <resource> subcommand, replacing prior separate scaffold commands.
- ›Adds URLReadTool for fetching and reading content from arbitrary URLs inside agent workflows.
- ›Adds app metadata to platform action tools, surfacing richer context for actions executed via the CrewAI platform.
1.15.11 Aug 5, 2026 · issue -014

CrewAI 1.15.11 adds an IBM Db2 search tool and enterprise account linking via project_id.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.11 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.11
```
- ›Adds project_id support to link open-source CrewAI usage to an enterprise account.
- ›Adds a new IBM Db2 search tool for querying Db2 databases from within a crew.
- ›Surfaces AMP in AGENTS.md and adds coding-agent detection to telemetry.
- ›Tracks interception-hook dispatches in telemetry for improved observability.
1.15.11 Aug 5, 2026 · issue 001

CrewAI 1.15.11 adds an IBM Db2 search tool, project_id for enterprise account linking, and AMP surfacing in AGENTS.md.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.11 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.11
```
- ›Adds project_id to link open-source CrewAI usage to an enterprise account.
- ›Adds IBM Db2 search tool, extending the library's built-in data-source integrations.
- ›Surfaces CrewAI AMP in AGENTS.md and adds coding-agent detection in telemetry.
- ›Tracks interception-hook dispatches in telemetry for improved observability of agent workflows.
1.15.10 Jul 31, 2026 · issue -019

CrewAI 1.15.10 adds skill usage event collection for agent observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.10 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.10
```
- ›Collects skill usage events, enabling observability into which agent skills are invoked during crew runs.
1.15.9 Jul 30, 2026 · issue -020

CrewAI 1.15.9 surfaces tool failures accurately, adds FlowFailedEvent, and introduces progressive disclosure for skills.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.9 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.9
```
- ›Emits FlowFailedEvent when a flow execution fails, enabling downstream handlers to react to flow-level errors.
- ›Surfaces tool failures as actual failures instead of silently reporting them as success, improving error visibility.
- ›Implements progressive disclosure for skills, controlling how skill details are revealed during execution.
1.15.9 Jul 30, 2026 · issue 001

CrewAI 1.15.9 adds FlowFailedEvent emission, surfaces tool failures accurately, and introduces progressive disclosure for skills.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.9 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.9
```
- ›Emits FlowFailedEvent when a flow execution fails, enabling event-driven error handling in Flows.
- ›Surfaces tool failures as actual failures instead of silently reporting them as success, giving agents accurate tool execution feedback.
- ›Adds progressive disclosure for skills, revealing skill detail incrementally rather than all at once.
1.15.8 Jul 28, 2026 · issue -022

CrewAI 1.15.8 adds WaitTool for pausing agent execution on long-running jobs.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.8 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.8
```
- ›Adds WaitTool to pause agent execution while waiting for long-running jobs to complete.
1.15.8 Jul 28, 2026 · issue 001

CrewAI 1.15.8 adds WaitTool for pausing agent execution during long-running jobs.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.8 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.8
```
- ›Adds WaitTool to the tool library, enabling agents to pause execution while waiting on long-running background jobs.
1.15.7 Jul 26, 2026 · issue -024

CrewAI 1.15.7 adds runtime skill usage event emission for observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.7 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.7
```
- ›Emits skill usage events at runtime for observability into agent behavior during crew execution.
1.15.7 Jul 26, 2026 · issue 001

CrewAI 1.15.7 emits skill usage events at runtime for agent observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.7 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.7
```
- ›Emits skill usage events at runtime, enabling observability of which skills agents invoke during crew execution.
1.15.7a1 Jul 26, 2026 · issue 001

CrewAI 1.15.7a1 adds runtime skill usage event emission for improved observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.7a1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.7a1
```
- ›Emits skill usage events at runtime, enabling observability pipelines to capture when and how skills are invoked by agents.
1.15.7a1 Jul 26, 2026 · issue -024

CrewAI 1.15.7a1 adds runtime skill usage event emission for improved observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.7a1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.7a1
```
- ›Emits skill usage events at runtime, enabling observability pipelines to capture when and how skills are invoked by agents.
1.15.5 Jul 20, 2026 · issue 001

CrewAI 1.15.5 adds authentication for skill registry downloads.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.5 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.5
```
- ›Adds authentication support for skill registry downloads, enabling access to protected or private skills from the CrewAI skill registry.
1.15.5 Jul 20, 2026 · issue -030

CrewAI 1.15.5 adds authentication for skill registry downloads.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.5 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.5
```
- ›Adds authentication support for skill registry downloads, enabling access to protected or private skills from the CrewAI skill registry.
1.15.3 Jul 16, 2026 · issue -034

CrewAI 1.15.3 adds step interception points, an @on hook dispatcher, and declarative flow support in the TUI.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.3 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.3
```
└──▷ USE IT
Intercept a specific execution boundary in a crew run to log or mutate state at that step.
python
```
from crewai import on

@on('after_llm_call')
def inspect_llm_output(context):
    print(context)
```
- ›Adds step interception points and a generic @on hook dispatcher for wiring execution-boundary hooks at precise points in agent and crew execution.
- ›Adds an organization_id parameter to the PlusAPI client for multi-org API targeting.
- ›Runs declarative flows on the TUI with a headless terminal fallback.
1.15.2 Jul 8, 2026 · issue -042

CrewAI 1.15.2 adds inline skill definitions, templated Flow action inputs, stream frame protocol, and AgentExecutor feedback handling.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.2 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.2
```
- ›Supports inline skill definitions, allowing skills to be defined directly without separate files.
- ›Adds templated Flow action inputs, enabling dynamic parameterization of Flow actions.
- ›Introduces a generated Flow Definition authoring skill for scaffolding flow definitions.
- ›Defines a stream frame protocol for flows, enabling structured streaming communication.
- ›Adds text helpers for Flow CEL prompts and flow skill examples.
+4 moreshow less
- ›Implements message setup and feedback handling in AgentExecutor.
- ›Adds repository agents to flow definitions.
- ›Types tool and app fields in CrewDefinition for stronger schema enforcement.
- ›Pulls latest LLM models dynamically in the crew wizard.
1.15.1 Jun 27, 2026 · issue -053

CrewAI 1.15.1 adds automatic Git init for new projects, enforces explicit crew definitions, and opens the deployment page post-deploy.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.1
```
- ›Automatically initializes a Git repository when generating a new CrewAI project.
- ›Requires explicit CrewAI project definitions, enforcing structured crew configuration at project creation.
- ›Opens the deployment page in the browser automatically after running a CLI deploy.
1.15.0 Jun 25, 2026 · issue -055

CrewAI 1.15.0 adds declarative Flow definitions, DMN mode, composite each actions, and full CLI TUI support for conversational flows.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.15.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.0
```
- ›Adds unified declarative flow loading so flows can be defined and loaded from config rather than pure code.
- ›Adds declarative Flow CLI support, enabling flow kickoff and management via CLI commands.
- ›Adds each composite action to FlowDefinition for iterating over steps within a flow.
- ›Adds optional if expression to each.do steps for conditional iteration logic inside flows.
- ›Adds single-agent actions directly inside Flow definitions.
+5 moreshow less
- ›Adds crew actions to FlowDefinition, letting flows orchestrate full crews as named actions.
- ›Adds inline crew definition loading so crews can be defined directly within a flow config.
- ›Implements DMN (Decision Model and Notation) mode support in crew creation and execution.
- ›Supports conversational flows in the CLI TUI for interactive, turn-based flow sessions.
- ›Tracks conversational flow turn usage in telemetry for observability into multi-turn interactions.
└──▷ BREAKING ON UPGRADE
- !StateProxy is removed from flow state access — code that uses StateProxy to read or write flow state will break.
1.14.7 Jun 11, 2026 · issue -069

CrewAI 1.14.7 adds a chat API for conversational flows, native Snowflake Cortex LLM, pluggable memory/RAG backends, and richer LLM event metadata.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.7 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.7
```
└──▷ USE IT
Enable conversational back-and-forth with a flow using the new chat API.
python
```
from crewai.flow.flow import Flow

flow = Flow()
response = flow.chat("What is the status of my order #1234?")
```
- ›Adds a chat API enabling conversational flows within CrewAI.
- ›Adds native Snowflake Cortex LLM provider integration.
- ›Introduces pluggable default backends for memory, knowledge, RAG, and flow — making the locking backend overridable.
- ›Surfaces real finish_reason, sampling parameters, and response.id on LLM events for richer observability.
- ›Types DSL triggers as route-aware decorators for more expressive flow definitions.
+3 moreshow less
- ›Builds FlowDefinition from Flow DSL metadata, separating DSL, definition, and runtime.
- ›Adds support for crew trained agents file, enabling persisted agent training state.
- ›Improves CrewAI import speed via lazy-loading of docling imports, unlocking faster cold starts.
1.14.6 May 28, 2026 · issue -083

CrewAI 1.14.6 adds env-var isolation in StdioTransport, Databricks env_var declarations, and enhanced planning configuration.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.6 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.6
```
- ›Adds env_vars declaration support on DatabricksQueryTool for explicit credential scoping.
- ›Improves planning configuration and observation handling in agent execution.
- ›Moves Skills Repository behind the CREWAI_EXPERIMENTAL gate as an experimental feature.
1.14.5 May 18, 2026 · issue -093

CrewAI 1.14.5 adds state-restore kickoff, ExaSearchTool highlights, and Daytona sandbox improvements.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.5 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.5
```
└──▷ USE IT
Resume a long-running Crew from a checkpoint state ID instead of restarting from scratch.
python
```
result = my_crew.kickoff(restore_from_state_id="<state_id>")
```
- ›Adds restore_from_state_id kickoff parameter to resume a Crew run from a previously saved state.
- ›Adds highlight results support to ExaSearchTool (formerly EXASearchTool).
- ›Improves Daytona sandbox tool integration.
- ›Defaults Crew agents to AgentExecutor, deprecating CrewAgentExecutor.
└──▷ BREAKING ON UPGRADE
- !The EXASearchTool is renamed to ExaSearchTool; imports using the old name will break.
1.14.4 Apr 30, 2026 · issue -111

CrewAI 1.14.4 adds custom persist keys, Azure OpenAI Responses API, Vertex AI workload identity, Tavily Research, and You.com MCP tools.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.4 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.4
```
- ›Supports custom persistence keys in the @persist decorator for fine-grained checkpoint control.
- ›Adds Responses API support for the Azure OpenAI provider.
- ›Forwards credential_scopes to the Azure AI Inference client for scoped authentication.
- ›Adds Vertex AI workload identity setup for credential-free GCP deployments.
- ›Adds Tavily Research and Get Research tools for AI-powered web research workflows.
+1 moreshow less
- ›Adds You.com MCP tools covering search, research, and content extraction.
1.14.3 Apr 24, 2026 · issue -117

CrewAI 1.14.3 adds lifecycle events for checkpoints, e2b/Daytona sandbox support, Bedrock V4, Azure DefaultAzureCredential fallback, and ~29% cold-start reduction.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.3 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.3
```
└──▷ USE IT
Use keyless Azure auth in environments where injecting an API key is not possible (e.g., managed identity on Azure).
python
```
from crewai import LLM

# No api_key provided — CrewAI falls back to DefaultAzureCredential automatically
llm = LLM(
    model="azure/<your-deployment>",
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    api_version="2024-02-01",
)
```
- ›Adds lifecycle events for checkpoint operations, enabling observability hooks around checkpoint create, resume, and fork actions.
- ›Supports e2b sandbox integration for secure code execution within crews.
- ›Adds Daytona sandbox tools for enhanced sandboxed functionality.
- ›Adds checkpoint and fork support to standalone agents, not just full crews.
- ›Falls back to DefaultAzureCredential when no API key is provided in the Azure integration, enabling keyless auth flows.
+2 moreshow less
- ›Adds Bedrock V4 support for AWS-hosted model access.
- ›Optimizes MCP SDK and event types to reduce cold start by ~29%, unlocking faster agent startup in latency-sensitive deployments.
1.14.2 Apr 17, 2026 · issue -124

CrewAI 1.14.2 adds checkpoint resume/fork/prune, template management, and enriched LLM token tracking.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.2 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.2
```
└──▷ USE IT
Resume a failed crew run from a checkpoint instead of restarting from scratch.
python
```
result = crew.kickoff(from_checkpoint=<checkpoint_id>)
```
- ›Adds checkpoint resume, diff, and prune commands for managing crew execution checkpoints.
- ›Adds from_checkpoint parameter to Agent.kickoff and related methods to restart runs from a saved state.
- ›Adds checkpoint forking with lineage tracking, enabling branched execution from any saved checkpoint.
- ›Adds template management commands for creating and managing project templates.
- ›Adds deploy validation CLI for pre-flight checks before deploying crews.
+1 moreshow less
- ›Enriches LLM token tracking with reasoning tokens and cache creation tokens.
1.14.1 Apr 8, 2026 · issue -133

CrewAI 1.14.1 adds async checkpoint TUI browser and proper resource management for streaming outputs.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.1
```
- ›Adds async checkpoint TUI browser for interactive inspection of crew checkpoints.
- ›Adds aclose()/close() methods and async context manager support to streaming outputs for proper resource cleanup.
1.14.0 Apr 7, 2026 · issue -134

CrewAI 1.14.0 adds runtime state checkpointing with SQLite storage, automatic checkpoint config, and new CLI commands for checkpoint management.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.14.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.0
```
└──▷ TRY IT
List all saved checkpoints for a crew run to find a resume point after an interrupted execution.
```
$ crewai checkpoint list
```
Inspect details of a specific checkpoint before deciding whether to resume from it.
```
$ crewai checkpoint info <checkpoint-id>
```
- ›Adds checkpoint list and checkpoint info CLI commands to inspect saved crew execution states.
- ›Introduces SqliteProvider for persistent checkpoint storage backend.
- ›Adds CheckpointConfig for configuring automatic checkpointing of crew runs.
- ›Implements runtime state checkpointing and an event system so long-running crews can be resumed after interruption.
- ›Adds guardrail_type and name fields to traces to distinguish guardrail events in observability pipelines.
+1 moreshow less
- ›Exports JsonProvider as an additional checkpoint storage option alongside SqliteProvider.
└──▷ BREAKING ON UPGRADE
- !CodeInterpreterTool is removed and code execution parameters are deprecated — any crew or agent config referencing CodeInterpreterTool or those parameters will break on upgrade.
1.13.0 Apr 2, 2026 · issue -139

CrewAI 1.13.0 adds unified runtime state serialization, richer telemetry spans, token usage events, and an A2UI extension.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.13.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.13.0
```
└──▷ USE IT
Capture token usage from every LLM call — useful for cost attribution or rate-limit monitoring in a pipeline.
python
```
from crewai.events import LLMCallCompletedEvent
from crewai.event_bus import event_bus

@event_bus.on(LLMCallCompletedEvent)
def on_llm_done(event: LLMCallCompletedEvent):
    print(event.token_usage)
```
- ›Adds RuntimeState RootModel for unified state serialization across crew runs.
- ›Enhances the event listener with new telemetry spans covering skill and memory events.
- ›Adds A2UI extension with v0.8/v0.9 support, schemas, and documentation.
- ›Emits token usage data in LLMCallCompletedEvent for per-call cost and usage tracking.
- ›Reduces framework overhead via a lazy event bus and skips tracing entirely when disabled.
1.12.1 Mar 26, 2026 · issue -146

CrewAI 1.12.1 adds Qdrant memory backend, agent skills, native OpenAI-compatible providers, and automatic hierarchical memory isolation.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.12.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.12.1
```
└──▷ USE IT
Connect CrewAI to a self-hosted Ollama instance as a native OpenAI-compatible provider, bypassing LiteLLM.
python
```
from crewai import LLM

llm = LLM(
    model="ollama/llama3",
    base_url="http://localhost:11434"
)
```
- ›Adds request_id to HumanFeedbackRequestedEvent for tracking human-in-the-loop feedback requests.
- ›Adds Qdrant Edge storage backend for the memory system.
- ›Adds agent skills support.
- ›Implements automatic root_scope for hierarchical memory isolation across agents.
- ›Supports native OpenAI-compatible providers: OpenRouter, DeepSeek, Ollama, vLLM, Cerebras, and Dashscope.
+2 moreshow less
- ›Adds logout command to the CLI.
- ›Adds docs-check command to analyze changes and generate docs with translations.
1.12.0 Mar 26, 2026 · issue -146

CrewAI 1.12.0 adds Qdrant memory backend, agent skills, native OpenAI-compatible providers, hierarchical memory isolation, and a CLI logout command.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.12.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.12.0
```
└──▷ TRY IT
Log out of the CrewAI platform from the terminal to rotate credentials or switch accounts.
```
$ crewai logout
```
- ›Adds Qdrant Edge as a storage backend for the memory system.
- ›Introduces agent skills as a new first-class capability for agents.
- ›Implements automatic root_scope for hierarchical memory isolation across crews.
- ›Supports native OpenAI-compatible providers: OpenRouter, DeepSeek, Ollama, vLLM, Cerebras, and Dashscope — without requiring LiteLLM.
- ›Adds a logout command to the CrewAI CLI.
+1 moreshow less
- ›Adds a docs-check CLI command to analyze changes and generate translated documentation.
1.11.1 Mar 23, 2026 · issue -149

CrewAI 1.11.1 adds flow_structure() serializer for programmatic Flow class introspection.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.11.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.11.1
```
└──▷ USE IT
Inspect the structure of a defined Flow programmatically — useful for debugging or generating visual representations of agent pipelines.
python
```
flow = MyFlow()
structure = flow.flow_structure()
print(structure)
```
- ›Adds flow_structure() serializer method to the Flow class for structured introspection of flow topology.
1.10.0 Feb 27, 2026 · issue -171

CrewAI 1.10.0 adds user input handling in Flows, enhanced MCP tool resolution, and auto-updating tool specs.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.10.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.10.0
```
- ›Adds user input handling in Flows, enabling human-in-the-loop data collection mid-flow execution.
- ›Enhances MCP tool resolution and emits related events for better observability of MCP-based tool use.
- ›Adds started_event_id tracking on the event bus for correlating flow and agent lifecycle events.
- ›Introduces auto-update of tools.specs, keeping tool specifications current without manual intervention.
- ›Enhances HITL (Human-in-the-Loop) self-loop functionality in human feedback integration.
+4 moreshow less
- ›Improves JSON argument parsing and validation in CrewAgentExecutor and BaseTool for more robust tool invocation.
- ›Migrates CLI HTTP client from requests to httpx, enabling async-compatible CLI operations.
- ›Adds versioned documentation support and yanked-version detection in release notes.
- ›Updates LanceDB version and adds lance-namespace packages for expanded vector store integration.
1.9.1 Jan 28, 2026 · issue -201

CrewAI 1.9.1 adds before/after tool call hooks and structured output support across providers
└──▷ GET THIS VERSION
```
$ git clone --branch 1.9.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.9.1
```
- ›Adds before and after tool call hooks in CrewAgentExecutor for intercepting and acting on tool invocations.
- ›Supports structured outputs and response_format parameter across LLM providers for typed agent responses.
1.9.0 Jan 27, 2026 · issue -202

CrewAI 1.9.0 adds A2A task execution, Keycloak SSO, structured outputs, and native multimodal file handling.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.9.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.9.0
```
└──▷ USE IT
Enforce a typed JSON schema on LLM output so downstream code can safely parse structured results.
python
```
from pydantic import BaseModel
from crewai import Agent, Task, Crew

class ThreatReport(BaseModel):
    severity: str
    affected_systems: list[str]
    recommendation: str

analyst = Agent(role="Threat Analyst", goal="Analyze security incidents", backstory="...", verbose=False)
task = Task(
    description="Analyze the attached incident log and produce a threat report.",
    expected_output="A structured threat report",
    agent=analyst,
    response_format=ThreatReport,
)
crew = Crew(agents=[analyst], tasks=[task])
result = crew.kickoff()
report: ThreatReport = result.pydantic
```
- ›Adds structured outputs and response_format support across providers for typed, schema-constrained LLM responses.
- ›Adds response_id field in streaming responses for tracking and correlating streamed completions.
- ›Adds event ordering and parent-child hierarchy to the event system for richer execution tracing.
- ›Adds Keycloak SSO provider support for enterprise authentication integration.
- ›Adds native multimodal file handling with OpenAI Responses API support.
+5 moreshow less
- ›Adds agent-to-agent (A2A) task execution utilities for orchestrating inter-agent workflows.
- ›Adds A2A server config and agent card generation for declaring and advertising agent capabilities.
- ›Adds additional A2A events with enriched event metadata for deeper observability into A2A interactions.
- ›Adds additional A2A transports for flexible agent-to-agent communication.
- ›Adds Galileo integration for LLM evaluation and observability.
1.8.1 Jan 15, 2026 · issue -214

CrewAI 1.8.1 adds Agent-to-Agent (A2A) task execution, server configuration, agent card generation, and new transports.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.8.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.8.1
```
- ›Adds A2A task execution utilities for orchestrating agent-to-agent workflows.
- ›Adds A2A server configuration and agent card generation to expose CrewAI agents as A2A-compatible services.
- ›Adds additional A2A transports, broadening connectivity options between agents.
- ›Adds Galileo to the integrations page, enabling Galileo observability for CrewAI workflows.
1.8.0 Jan 8, 2026 · issue -221

CrewAI 1.8.0 adds Agent-to-Agent async/streaming/push update mechanisms and Human-in-the-Loop support for Flows.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.8.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.8.0
```
- ›Adds native async chain support for agent-to-agent (A2A) communication.
- ›Introduces A2A update mechanisms — poll, stream, and push — with configurable handlers.
- ›Adds Human-in-the-Loop (HITL) feedback support directly within Flows via global flow configuration.
- ›Adds streaming tool call events for real-time observability of tool execution.
- ›Introduces production-ready Flows and Crews architecture.
+1 moreshow less
- ›Improves EventListener and TraceCollectionListener for enhanced event handling.
1.7.1 Dec 16, 2025 · issue -244

CrewAI 1.7.1 adds a --no-commit flag to the bump command and switches to JSON schema for tool argument serialization.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.7.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.7.1
```
└──▷ TRY IT
Bump the project version without creating a git commit, useful in CI pipelines where you control commits separately.
```
$ crewai bump --no-commit
```
- ›Adds --no-commit flag to the bump command, allowing version bumps without auto-committing changes.
- ›Switches tool argument serialization to use JSON schema for more structured, interoperable tool definitions.
1.7.0 Dec 9, 2025 · issue -251

CrewAI 1.7.0 adds comprehensive async support across flows, crews, tasks, tools, memory, knowledge, and LLMs.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.7.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.7.0
```
└──▷ USE IT
Run a full crew asynchronously to avoid blocking the event loop in an async application or service.
python
```
import asyncio
from crewai import Crew, Agent, Task

agent = Agent(role="Analyst", goal="Analyze data", backstory="Expert analyst")
task = Task(description="Summarize the report", agent=agent, expected_output="Summary")
crew = Crew(agents=[agent], tasks=[task])

async def main():
    result = await crew.kickoff_async()
    print(result)

asyncio.run(main())
```
Kick off a Flow asynchronously so multiple flows can run concurrently in the same process.
python
```
import asyncio
from crewai.flow.flow import Flow, start

class MyFlow(Flow):
    @start()
    async def run_step(self):
        return "done"

async def main():
    flow = MyFlow()
    result = await flow.kickoff_async()
    print(result)

asyncio.run(main())
```
- ›Adds async flow kickoff, enabling non-blocking orchestration of multi-agent pipelines.
- ›Adds async crew support so entire crews can be run concurrently without blocking the event loop.
- ›Adds async task support for individual task execution within crews.
- ›Adds async knowledge and memory support for non-blocking retrieval and storage operations.
- ›Adds async support for tools and the agent executor with improved typing.
+4 moreshow less
- ›Adds native async tool support, allowing tools to be defined and invoked asynchronously.
- ›Adds async LLM support for non-blocking model inference calls.
- ›Implements the agent-to-agent (A2A) extensions API with async agent card caching.
- ›Introduces system event types and a handler for observability and lifecycle events.
1.6.0 Nov 25, 2025 · issue -265

CrewAI 1.6.0 adds streaming results for flows and crews, Entra ID CLI login, a Merge Agent Handler tool, and Gemini 2.5 Pro preview support.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.6.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.6.0
```
- ›Adds streaming result support to flows and crews, enabling real-time output consumption.
- ›Supports CLI login with Entra ID (Microsoft Entra / Azure AD) for enterprise SSO authentication.
- ›Adds Merge Agent Handler tool for combining agent outputs.
- ›Enhances flow event state management for improved control over flow lifecycle events.
1.5.0 Nov 16, 2025 · issue -274

CrewAI 1.5.0 adds LLM call hooks, exposes messages to task outputs, and improves A2A trust and Okta data.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.5.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.5.0
```
└──▷ USE IT
Inspect the full message history attached to a task result for auditing or downstream processing.
python
```
result = crew.kickoff()
for task_output in result.tasks_output:
    print(task_output.messages)
```
- ›Adds before and after LLM call hooks in CrewAgentExecutor for observability and control around every LLM invocation.
- ›Exposes messages to TaskOutput and LiteAgentOutputs, giving callers access to the full message history from a task run.
- ›Adds A2A trust remote completion status flag for agent-to-agent trust workflows.
- ›Fetches and stores richer data about Okta authorization servers for enhanced auth integrations.
- ›Enhances schema description of QdrantVectorSearchTool for improved usability and discoverability.
1.4.0 Nov 7, 2025 · issue -283

CrewAI 1.4.0 adds first-class MCP support, LLM message interceptor hooks, and enhances the QdrantVectorSearchTool.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.4.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.4.0
```
- ›Adds first-class support for MCP (Model Context Protocol) with improved concurrent tool execution handling.
- ›Enhances QdrantVectorSearchTool with new capabilities for vector search workflows.
- ›Adds support for non-AST plot routes, expanding flow visualization options.
- ›Caches i18n prompts for more efficient repeated use across agents.
1.3.0 Nov 1, 2025 · issue -289

CrewAI 1.3.0 enhances the QdrantVectorSearchTool with improved vector search capabilities.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.3.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.3.0
```
- ›Enhances QdrantVectorSearchTool with expanded capabilities for vector-based semantic search.
1.2.1 Oct 27, 2025 · issue -294

CrewAI 1.2.1 adds Datadog observability integration and MCP/app support in LiteAgent.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.2.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.2.1
```
- ›Adds Datadog integration for observability and tracing of CrewAI agent runs.
- ›Enables apps and MCPs (Model Context Protocol servers) as supported resources in LiteAgent.
1.1.0 Oct 21, 2025 · issue -300

CrewAI 1.1.0 adds multi-provider LLM support in InternalInstructor, a mypy plugin base, and improves QdrantVectorSearchTool.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.1.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.1.0
```
- ›Adds multi-provider LLM support to InternalInstructor, enabling structured output extraction across different LLM backends.
- ›Introduces a mypy plugin base for improved static type checking of CrewAI code.
- ›Improves QdrantVectorSearchTool for better vector search integration.
1.0.0 Oct 20, 2025 · issue -301

CrewAI 1.0.0 adds enhanced knowledge/guardrail event handling and tool repository credential injection for the run command.
└──▷ GET THIS VERSION
```
$ git clone --branch 1.0.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.0.0
```
- ›Enhances knowledge and guardrail event handling in the Agent class for more robust agent lifecycle control.
- ›Injects tool repository credentials automatically in the crewai run command, enabling authenticated tool source access.
0.201.1 Sep 26, 2025 · issue -324

CrewAI 0.201.1 adds ChromaDB support for WatsonX and VoyageAI embedding providers.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.201.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.201.1
```
- ›Adds ChromaDB compatibility for watsonx and voyageai embedding providers.
└──▷ BREAKING ON UPGRADE
- !The watson embedding provider is renamed to watsonx; environment variable prefixes are updated accordingly — any config using the old watson provider name or its env var prefixes will break on upgrade.
0.201.0 Sep 26, 2025 · issue -324

CrewAI 0.201.0 adds a crewai uv CLI wrapper, custom embedding types, thread-safe platform context, and failed-trace marking for observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.201.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.201.0
```
└──▷ USE IT
Set batch_size on an embedder to avoid hitting token limits when indexing large document sets.
python
```
embedder_config = {
    "provider": "openai",
    "config": {
        "model": "text-embedding-3-small",
        "batch_size": 50
    }
}
```
- ›Adds crewai uv wrapper command to invoke uv directly through the CrewAI CLI.
- ›Supports batch_size configuration on embedders to stay within token limits.
- ›Introduces thread-safe platform context management for concurrent crew workflows.
- ›Enables marking traces as failed in observability workflows.
- ›Adds custom embedding types and provider migration support.
+3 moreshow less
- ›Upgrades ChromaDB dependency to v1.1.0 with type improvements.
- ›Introduces Pydantic v2 support along with pydantic-settings and dependency group reorganization.
- ›Adds Pydantic-compatible import validation, replacing legacy utilities.
0.193.0 Sep 19, 2025 · issue -331

CrewAI 0.193.0 makes RAG/knowledge/memory search params fully configurable and enables ChromaDB OpenAI embeddings.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.193.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.193.0
```
- ›Makes search parameters for RAG, knowledge, and memory fully configurable.
- ›Enables ChromaDB to use OpenAI API as an embedding function.
- ›Adds deeper observability tools for user-level insights.
- ›Introduces thread-safe platform context management.
- ›Unifies RAG storage system with instance-specific client support.
+2 moreshow less
- ›Adds ephemeral trace improvements for better trace control.
- ›Adds missing event exports to __init__.py for consistent module behavior.
0.186.0 Sep 10, 2025 · issue -340

CrewAI 0.186.0 adds Qdrant RAG support, partial flow resumability, auto-injection of trigger payloads, and a new config reset command.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.186.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.186.0
```
└──▷ TRY IT
Reset stored auth/API tokens without being prompted for email — useful when rotating credentials in CI.
```
$ crewai config reset
```
- ›Adds support for crewai_trigger_payload auto-injection into crew/flow executions.
- ›Introduces a centralized RAG configuration system with optional imports, including Qdrant as a new RAG provider.
- ›Implements generic clients for ChromaDB and Qdrant storage backends.
- ›Enables partial flow resumability, allowing interrupted flows to resume mid-execution.
- ›Displays task names in verbose output for clearer runtime observability.
+3 moreshow less
- ›Introduces crewai config reset command for resetting authentication tokens.
- ›Adds additional parameters to Flow.start() methods for more flexible flow invocation.
- ›Introduces centralized configuration for embedding types and a base embedding client.
└──▷ BREAKING ON UPGRADE
- !A deprecation warning is now emitted for Task.max_retries; callers using this field should migrate before it is removed.
0.175.0 Aug 28, 2025 · issue -353

CrewAI 0.175.0 adds Qdrant RAG support, crewai config reset, auto-injected trigger payloads, and a centralized embedding client system.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.175.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.175.0
```
└──▷ TRY IT
Clear cached login tokens from the local crewAI config when rotating credentials or switching accounts.
```
$ crewai config reset
```
- ›Adds Qdrant as a RAG provider, with a new generic Qdrant client alongside an existing ChromaDB client.
- ›Introduces a centralized embedding types system and a base embedding client for standardized embedding configuration.
- ›Adds crewai config reset command to clear stored tokens.
- ›Enables crewai_trigger_payload auto-injection into flows triggered by automation.
- ›Adds support for additional parameters in Flow.start() methods.
+3 moreshow less
- ›Displays task names in verbose CLI output for better observability.
- ›Simplifies RAG client initialization with a new RAG configuration system.
- ›Adds support to remove Auth0 and email entry on crewai login.
0.165.1 Aug 19, 2025 · issue -362

CrewAI 0.165.1 adds agent message history to ExternalMemory, auto-injects trigger payloads, and links memory entries to agent IDs in Mem0.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.165.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.165.1
```
- ›Includes exchanged agent messages in ExternalMemory metadata for richer memory context.
- ›Automatically injects crewai_trigger_payload into crew execution context.
- ›Adds support for agent_id-linked memory entries in Mem0 integration.
└──▷ BREAKING ON UPGRADE
- !The internal flag inject_trigger_input is renamed to allow_crewai_trigger_context; any code or config referencing the old name will break.
- !The AgentOps integration has been removed; any setup relying on it will no longer function.
0.165.0 Aug 19, 2025 · issue -362

CrewAI 0.165.0 enriches ExternalMemory with agent messages, auto-injects trigger payloads, and adds agent_id-linked Mem0 entries.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.165.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.165.0
```
- ›Includes exchanged agent messages in ExternalMemory metadata for richer memory context.
- ›Automatically injects crewai_trigger_payload into crew runs triggered externally.
- ›Adds support for agent_id-linked memory entries in Mem0 integration.
└──▷ BREAKING ON UPGRADE
- !The internal flag inject_trigger_input is renamed to allow_crewai_trigger_context; any configuration or code referencing inject_trigger_input will break.
- !The AgentOps integration has been removed; any setup relying on it will stop working.
0.159.0 Aug 13, 2025 · issue -363

CrewAI 0.159.0 adds enterprise CLI setup command and partial flow resumability support.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.159.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.159.0
```
└──▷ TRY IT
Configure enterprise credentials and parameters without manually editing config files.
```
$ crewai enterprise configure
```
- ›Adds enterprise configure CLI command for streamlined enterprise setup.
- ›Introduces partial flow resumability, enabling flows to resume from an intermediate state rather than restarting entirely.
0.157.0 Aug 6, 2025 · issue -363

CrewAI 0.157.0 adds a crewai config CLI command group, Okta device authorization, LangDB integration, and initial tracing capabilities.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.157.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.157.0
```
└──▷ TRY IT
Inspect or set CrewAI CLI configuration without manually editing config files.
```
$ crewai config
```
- ›Adds crewai config CLI command group for managing CLI configuration.
- ›Supports device authorization with Okta for authenticated flows.
- ›Adds LangDB integration for observability and query tracing.
- ›Introduces initial tracing capabilities for crew and flow execution.
- ›Enables persisting Flow state with BaseModel entries.
+1 moreshow less
- ›Adds default value support for crew.name.
└──▷ BREAKING ON UPGRADE
- !Support for the deprecated User Memory system has been dropped; any setup relying on it will stop working after upgrading.
0.152.0 Jul 30, 2025 · issue -364

CrewAI 0.152.0 adds custom Flow names, a dedicated RAG module, and timezone-aware event timestamps.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.152.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.152.0
```
- ›Supports custom flow names on the Flow class for clearer identification of flows.
- ›Refactors RAG components into a dedicated top-level module for cleaner imports and organization.
- ›Adds timezone support to event timestamps for accurate time-based event tracking.
0.150.0 Jul 23, 2025 · issue -364

CrewAI 0.150.0 adds ad-hoc tool calling, Mem0 v2 storage, SerperScrapeWebsiteTool, and Bedrock AgentCore browser/code interpreter toolkits.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.150.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.150.0
```
- ›Adds support for ad-hoc tool calling via the internal LLM class, enabling on-demand tool invocation outside of standard crew/agent flows.
- ›Upgrades Mem0 Storage integration from v1.1 to v2.
- ›New SerperScrapeWebsiteTool extracts clean content from URLs using Serper.
- ›Integrates Bedrock AgentCore browser and code interpreter toolkits for use with Bedrock agents.
- ›Adds UserMemory deprecation notice to signal future removal.
0.148.0 Jul 16, 2025 · issue -364

CrewAI 0.148.0 introduces Agent evaluation functionality with thread-safe AgentEvaluator and neatlogs integration.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.148.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.148.0
```
- ›Introduces Agent evaluation functionality with AgentEvaluator, supporting regression testing and experiment methods for both Agent and LiteAgent.
- ›Enables event emission during Agent evaluation for observability into evaluation runs.
- ›Adds crew context tracking for LLM guardrail events.
- ›Adds integration with neatlogs for structured agent log management.
0.141.0 Jul 9, 2025 · issue -364

CrewAI 0.141.0 adds crew context tracking for LLM guardrail events.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.141.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.141.0
```
- ›Adds crew context tracking for LLM guardrail events, enabling richer auditability of guardrail decisions within a crew's execution context.
0.140.0 Jul 2, 2025 · issue -364

CrewAI 0.140.0 adds LLM call tracking by task/agent, MemoryEvents monitoring, and a WorkOS CLI login command.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.140.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.140.0
```
- ›Tracks LLM calls broken down by task and agent for observability into crew execution costs and patterns.
- ›Introduces MemoryEvents to monitor memory usage within a crew run.
- ›Adds console logging for memory system and LLM guardrail events.
- ›Improves data training support for models up to 7B parameters.
- ›Adds workos login command to the CLI for WorkOS-based authentication.
0.134.0 Jun 25, 2025 · issue -365

CrewAI 0.134.0 adds MCP multi-tool agent support, Tool-attribute initialization, and Oxylabs web scraping tools.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.134.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.134.0
```
- ›Supports initializing a tool directly from defined Tool attributes, enabling programmatic tool construction.
- ›Adds an official way to use MCP Tools within a CrewBase class.
- ›Enhances MCP tools support to allow selecting multiple tools per agent inside CrewBase.
- ›Adds Oxylabs Web Scraping tools as a built-in integration.
0.130.0 Jun 12, 2025 · issue -365

CrewAI 0.130.0 adds LiteAgent with Guardrail integration, async tool execution, and multi-org CLI support.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.130.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.130.0
```
- ›Introduces LiteAgent with built-in Guardrail integration for lightweight, constrained agent workflows.
- ›Enables async tool execution for more efficient, non-blocking agent workflows.
- ›Adds support for multi-org actions in the CLI.
- ›Upgrades LiteLLM to support the latest OpenAI version.
0.126.0 Jun 5, 2025 · issue -365

CrewAI 0.126.0 adds Python 3.13 support, streamable-HTTP MCP transport, prompt/memory transparency, and tool-logging.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.126.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.126.0
```
- ›Adds Python 3.13 support.
- ›Adds streamable-HTTP transport support in MCP integration.
- ›Enables tools to be loaded from an Agent repository via their own module.
- ›Persists available tools from a Tool repository across sessions.
- ›Logs tool usage when called by an LLM for observability.
+2 moreshow less
- ›Introduces transparency features for prompts and memory systems.
- ›Adds community analytics support.
0.121.0 May 22, 2025 · issue -366

CrewAI 0.121.0 adds markdown rendering for Tasks, reasoning for Agents, automatic date injection, and a HallucinationGuardrail.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.121.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.121.0
```
└──▷ USE IT
Enable markdown output for a task so results are returned as formatted markdown.
python
```
from crewai import Task

task = Task(
    description="Summarize the latest threat intelligence report.",
    expected_output="A structured summary of key findings.",
    markdown=True
)
```
Enable reasoning on an agent and inject today's date automatically for time-sensitive analysis workflows.
python
```
from crewai import Agent

analyst = Agent(
    role="Threat Analyst",
    goal="Identify emerging threats from recent feeds.",
    backstory="Expert in cyber threat intelligence.",
    reasoning=True,
    inject_date=True
)
```
- ›Adds markdown attribute to the Task class for controlling markdown-formatted output.
- ›Adds reasoning attribute to the Agent class to enable or configure agent reasoning behavior.
- ›Adds inject_date flag to Agent for automatic date injection into agent context.
- ›Implements HallucinationGuardrail for detecting and guarding against hallucinated outputs.
0.120.0 May 14, 2025 · issue -366

CrewAI 0.120.0 adds agent-from-repository loading, empty Task context, and direct knowledge initialization.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.120.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.120.0
```
- ›Supports loading an Agent directly from a repository.
- ›Enables setting an empty context for a Task.
- ›Introduces direct initialization of knowledge, bypassing knowledge_sources.
0.119.0 May 8, 2025 · issue -366

CrewAI 0.119.0 adds parent flow identification for Crew and LiteAgent and knowledge retrieval prompt rewriting in Agent.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.119.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.119.0
```
- ›Enables parent flow identification for Crew and LiteAgent, improving traceability in nested flow architectures.
- ›Introduces knowledge retrieval prompt rewriting in Agent for improved tracking and debugging of RAG-based workflows.
0.118.0 Apr 30, 2025 · issue -367

CrewAI 0.118.0 adds no-code Guardrail creation and renames TaskGuardrail to LLMGuardrail.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.118.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.118.0
```
- ›Adds support for no-code Guardrail creation to simplify AI behavior controls without writing custom guardrail logic.
└──▷ BREAKING ON UPGRADE
- !TaskGuardrail is renamed to LLMGuardrail; any code importing or referencing TaskGuardrail will break on upgrade.
0.117.0 Apr 28, 2025 · issue -367

CrewAI 0.117.0 adds result_as_answer decorator support, GPT-4.1/Gemini-2.x models, and a HuggingFace CLI provider.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.117.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.117.0
```
└──▷ USE IT
Return a tool's output directly as the agent's final answer, bypassing further LLM reasoning — useful for deterministic lookup tools where you trust the result completely.
python
```
from crewai.tools import tool

@tool(result_as_answer=True)
def lookup_cve(cve_id: str) -> str:
    """Fetch CVE details from internal database."""
    return fetch_cve_record(cve_id)
```
- ›Adds result_as_answer parameter to the @tool decorator, allowing a tool's output to be used directly as the agent's final answer.
- ›Supports new language models: GPT-4.1, Gemini-2.0, and Gemini-2.5 Pro.
- ›Adds HuggingFace as a provider option in the CrewAI CLI.
- ›Enhances knowledge management capabilities.
0.114.0 Apr 10, 2025 · issue -367

CrewAI 0.114.0 lets agents run standalone, adds custom LLM support, external memory, and Opik observability.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.114.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.114.0
```
└──▷ USE IT
Run a single agent as a standalone unit — useful in Flows or one-off tasks without assembling a full Crew.
python
```
from crewai import Agent

researcher = Agent(
    role="Research Analyst",
    goal="Find the latest CVEs for Apache HTTP Server",
    backstory="You are an expert in vulnerability research.",
    llm="gpt-4o"
)

result = researcher.kickoff()
print(result)
```
- ›Enables agents as atomic, standalone units — call Agent(...).kickoff() without a full Crew.
- ›Supports custom LLM implementations for bringing your own model client.
- ›Integrates External Memory for persistent agent knowledge across runs.
- ›Adds Opik observability integration for tracing and monitoring agent workflows.
- ›Adds wildcard support to emit() for flexible event broadcasting.
+3 moreshow less
- ›Introduces secure fingerprints for agents and crews to uniquely identify and track them.
- ›Adds multimodal agent validation to enforce correct configuration of multimodal agents.
- ›Enhanced YAML extraction for more robust crew and agent definition parsing.
0.108.0 Mar 17, 2025 · issue -368

CrewAI 0.108.0 adds agent fingerprints, richer event listener visualization, and improved LLM streaming event handling.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.108.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.108.0
```
- ›Adds fingerprints to agents/tasks for unique identity tracking across runs.
- ›Enhances LLM streaming response handling with an improved event system for real-time observability.
- ›Enriches the event listener with rich visualization and improved logging output.
- ›Includes model_name in relevant model representations for better introspection.
0.105.0 Mar 9, 2025 · issue -368

CrewAI 0.105.0 adds Flow state export, an event emitter for LLM observability, multi-router support, and Python 3.10 compatibility.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.105.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.105.0
```
- ›Adds Flow state export and improved state utilities for inspecting and persisting Flow execution state.
- ›Introduces an event emitter for observability, enabling tracking of LLM calls and agent events.
- ›Supports multiple router calls within a single Flow, enabling more complex routing logic.
- ›Adds support for Python 3.10.
- ›Adds ChatOllama integration via langchain_ollama for local LLM usage.
+3 moreshow less
- ›Adds context window size support for the o3-mini model.
- ›Enhances agent knowledge setup with an optional crew-level embedder configuration.
- ›Adds QdrantVectorSearchTool guide and event listener usage documentation.
0.102.0 Feb 13, 2025 · issue -369

CrewAI 0.102.0 adds QdrantVectorSearchTool, JSON logging, multi-tab Excel knowledge, and custom embedder support.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.102.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.102.0
```
└──▷ USE IT
Search a Qdrant vector collection from within a CrewAI agent tool to ground responses in your own embeddings.
python
```
from crewai_tools import QdrantVectorSearchTool

tool = QdrantVectorSearchTool(
    collection_name="security-advisories",
    url="http://localhost:6333",
    api_key="<your-qdrant-api-key>"
)
agent = Agent(role="Threat Analyst", tools=[tool], ...)
```
- ›Adds QdrantVectorSearchTool for vector similarity search against Qdrant collections.
- ›Supports JSON format for logging output, improving observability pipeline integration.
- ›Enables multi-tab Excel file processing in excel_knowledge_source.py for richer knowledge ingestion.
- ›Adds a reset_memories function to the Crew class for programmatic memory management.
- ›Supports custom embedder configuration for enhanced embedding setup in knowledge sources.
+1 moreshow less
- ›Integrates MLflow tracing support for agent execution observability.
0.100.0 Jan 28, 2025 · issue -370

CrewAI 0.100.0 adds Amazon SageMaker as a supported LLM provider.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.100.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.100.0
```
- ›Supports Amazon SageMaker as an LLM provider for running agents against hosted SageMaker endpoints.
0.98.0 Jan 20, 2025 · issue -370

CrewAI 0.98.0 adds Conversation Crew, flow state persistence with @persist, and SambaNova/NVIDIA NIM/VoyageAI integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.98.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.98.0
```
└──▷ USE IT
Persist flow state across runs so a long-running security workflow can resume where it left off after interruption.
python
```
from crewai.flow.persistence import persist, FlowPersistence

class MySecurityFlow(Flow):
    @persist
    def analyze_targets(self):
        # state is automatically saved after this method completes
        ...
```
- ›New Conversation Crew v1 enables interactive, dialogue-driven crew execution.
- ›Adds unique IDs to flow states for tracking and referencing individual flow runs.
- ›New @persist decorator with FlowPersistence interface enables durable flow state persistence across runs.
- ›Adds SambaNova as a new LLM provider integration.
- ›Adds NVIDIA NIM as a new provider via the CrewAI CLI.
+1 moreshow less
- ›Introduces VoyageAI as a new embedding/model integration.
0.95.0 Jan 4, 2025 · issue -370

CrewAI 0.95.0 adds multimodal agents, programmatic guardrails, multi-round HITL, Gemini 2.0, Langfuse, Portkey, Docling, and Weaviate support.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.95.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.95.0
```
- ›Adds multimodal abilities to Crew, enabling agents to process image and other non-text inputs.
- ›Introduces programmatic guardrails for enforcing constraints on agent outputs at runtime.
- ›Supports multiple rounds of Human-in-the-Loop (HITL) interaction within a single crew run.
- ›Adds Gemini 2.0 model support.
- ›Delivers CrewAI Flows improvements for more capable workflow orchestration.
+6 moreshow less
- ›Adds workflow permissions to control agent and task access within a crew.
- ›Supports Langfuse observability via LiteLLM integration.
- ›Adds Portkey integration for LLM gateway and observability.
- ›Introduces interpolate_only method on prompt/template handling for targeted variable substitution.
- ›Adds Docling support for document ingestion and parsing.
- ›Adds Weaviate support as a vector store for agent knowledge.
0.86.0 Dec 5, 2024 · issue -371

CrewAI 0.86.0 adds multi-round Human-in-the-Loop follow-up and expanded knowledge tooling
└──▷ GET THIS VERSION
```
$ git clone --branch 0.86.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.86.0
```
- ›Supports multiple rounds of Human-in-the-Loop (HITL) follow-up, enabling iterative human feedback within a crew's execution flow.
- ›Adds Nvidia NIM as a supported provider for custom LLM configuration.
- ›Introduces a knowledge demo and improved knowledge documentation to help practitioners integrate knowledge sources into crews.
└──▷ BREAKING ON UPGRADE
- !All references to Pipeline and PipelineRouter have been removed; any working setup that uses these constructs will break on upgrade.
0.85.0 Dec 4, 2024 · issue -371

CrewAI 0.85.0 adds agent-level knowledge, removes LangChain dependency, and improves typed task outputs.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.85.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.85.0
```
- ›Adds knowledge support at the individual agent level, enabling per-agent knowledge bases alongside crew-level knowledge.
- ›Removes LangChain as a dependency, reducing the library's footprint and eliminating LangChain version conflicts.
- ›Improves typed task outputs for stronger, more predictable structured results from tasks.
- ›Adds Tool Repository authentication via crewai login, enabling access to hosted tools.
└──▷ BREAKING ON UPGRADE
- !LangChain has been removed as a dependency; any code or configuration that imports or relies on LangChain internals through CrewAI will break on upgrade.
v0.83.0 Nov 25, 2024 · issue -372

CrewAI v0.83.0 adds crew lifecycle callbacks, agent knowledge pre-seeding, and Mem0 memory/preference retrieval.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.83.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.83.0
```
└──▷ USE IT
Run setup or teardown logic around a crew execution using the new lifecycle hooks.
python
```
@crew
def my_crew(self) -> Crew:
    return Crew(
        agents=self.agents,
        tasks=self.tasks,
        before_kickoff=self.before_kickoff,
        after_kickoff=self.after_kickoff,
    )
```
- ›Adds before_kickoff and after_kickoff crew callbacks for hooking into crew lifecycle events.
- ›Supports pre-seeding agents with Knowledge so agents start with domain context before execution begins.
- ›Adds Mem0 integration for retrieving user preferences and memories during agent runs.
0.79.0 Nov 10, 2024 · issue -372

CrewAI 0.79.0 adds flow inputs, IBM Watson memory integration, and broader log storage data type support.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.79.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.79.0
```
└──▷ USE IT
Build a custom tool using the now top-level BaseTool import rather than reaching into submodules.
python
```
from crewai import BaseTool

class MyScanner(BaseTool):
    name: str = "Port Scanner"
    description: str = "Scans open ports on a given host."

    def _run(self, host: str) -> str:
        # tool logic here
        return f"Scanning {host}"
```
- ›Adds inputs parameter support to flows, enabling dynamic data to be passed into flow execution.
- ›Enhances log storage to support a wider range of data types beyond strings.
- ›Moves BaseTool to the main package and centralizes tool description generation, simplifying custom tool authoring.
- ›Raises an explicit error when an LLM returns no response, making silent failures visible.
0.76.9 Oct 30, 2024 · issue -373

CrewAI 0.76.9 adds crewai flow add-crew and updates the flow plot command to crewai flow plot.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.76.9 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.76.9
```
└──▷ TRY IT
Add a new crew to an existing flow without manually editing boilerplate — useful when extending a flow with a parallel or sequential crew.
```
$ crewai flow add-crew
```
Visualize your flow's structure to review crew connections and routing logic before deploying.
```
$ crewai flow plot
```
- ›New crewai flow add-crew CLI command lets you scaffold and add additional crews to an existing flow.
- ›Flow visualization command updated to crewai flow plot.
- ›Forwards install command options to uv sync for more flexible dependency management.
- ›Python 3.10 support added via tomli dependency.
└──▷ BREAKING ON UPGRADE
- !The flow plot command is renamed from its previous form to crewai flow plot; any scripts or docs referencing the old command will break.
0.76.0 Oct 23, 2024 · issue -373

CrewAI 0.76.0 adds unsafe code execution support with Docker install and runtime checks.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.76.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.76.0
```
- ›Supports unsafe code execution with Docker install and running checks for sandboxed agent workflows.
0.74.0 Oct 18, 2024 · issue -373

CrewAI 0.74.0 adds model selection and API key submission via CLI, a new Memory Base, and migrates tooling to UV.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.74.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.74.0
```
- ›Migrates the project toolchain from Poetry to UV, with CLI adapted for UV workflows.
- ›New CLI capability to select models and submit API keys directly from the command line.
- ›Introduces a new Memory Base foundation for agent memory management.
└──▷ BREAKING ON UPGRADE
- !Project toolchain has migrated from Poetry to UV; existing Poetry-based setups will require migration to continue working.
0.70.1 Oct 11, 2024 · issue -373

CrewAI 0.70.1 adds Flows with a visual debugger, plus new CLI commands for scaffolding flows and tools.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.70.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.70.1
```
└──▷ TRY IT
Scaffold a new Flow project to start building multi-step agentic pipelines.
```
$ crewai create flow
```
Scaffold a new custom tool ready for implementation and publishing.
```
$ crewai tool create <tool>
```
- ›New Flow feature enables structured, multi-step agentic pipelines.
- ›Flow visualizer lets practitioners inspect and debug Flow execution graphs.
- ›New crewai create flow command scaffolds a Flow project from the CLI.
- ›New crewai tool create <tool> command scaffolds a custom tool from the CLI.
- ›Adds Git validations when publishing tools to enforce clean repository state.
0.64.0 Sep 27, 2024 · issue -374

CrewAI 0.64.0 introduces an initial Tools API and raises the default max iterations to 20.
└──▷ GET THIS VERSION
```
$ git clone --branch 0.64.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 0.64.0
```
- ›Adds an initial Tools API for programmatic tool management.
- ›Increases default max iterations from its previous limit to 20, enabling more complex agent reasoning loops out of the box.
v0.63.0 Sep 24, 2024 · issue -374

CrewAI v0.63.0 adds a unified LLM class, custom memory interfaces, and switches the default model to GPT-4o-mini.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.63.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.63.0
```
└──▷ USE IT
Instantiate an agent with the new LLM class to explicitly target a specific model via LiteLLM, decoupling model selection from environment defaults.
python
```
from crewai import Agent, LLM

llm = LLM(model="gpt-4o")
agent = Agent(
    role="Security Analyst",
    goal="Identify vulnerabilities in the provided code.",
    backstory="Expert in application security.",
    llm=llm
)
```
- ›New LLM class provides a unified interface for interacting with language models via LiteLLM.
- ›Supports custom memory interfaces, enabling practitioners to plug in their own memory backends.
- ›Changes the default model to GPT-4o-mini.
└──▷ BREAKING ON UPGRADE
- !The default model is now GPT-4o-mini; crews that relied on the previous default model will use GPT-4o-mini after upgrading unless an explicit model is set.
v0.60.0 Sep 16, 2024 · issue -374

CrewAI v0.60.0 drops LangChain, rebuilds the executor, and adds o1-model support with new agent controls.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.60.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.60.0
```
└──▷ USE IT
Use CrewAI with an o1-series model that doesn't support system prompts or stop words.
python
```
from crewai import Agent

agent = Agent(
    role="Analyst",
    goal="Analyze the dataset",
    backstory="You are a data expert.",
    llm="o1-preview",
    use_system_prompt=False,
    use_stop_words=False
)
```
Cap API usage and prevent runaway agent loops by setting rate limits and a max iteration count.
python
```
from crewai import Agent

agent = Agent(
    role="Researcher",
    goal="Find key insights",
    backstory="You are a senior researcher.",
    max_rpm=10,
    max_iter=5
)
```
- ›Removes LangChain dependency with a fully rebuilt internal executor for improved reliability.
- ›New use_system_prompt flag on Agent lets you disable system prompts for models that don't support them.
- ›New use_stop_words flag on Agent allows disabling stop words to support o1-series models.
- ›Adds configurable max requests per minute at the crew/agent level.
- ›Adds a configurable maximum number of iterations before an agent is forced to produce a final answer.
+3 moreshow less
- ›New token calculation flow for accurate usage tracking across agent runs.
- ›New logging of crew and agent execution for improved observability.
- ›sliding_context_window is renamed to respect_context_window and is now enabled by default.
└──▷ BREAKING ON UPGRADE
- !The sliding_context_window setting is renamed to respect_context_window; any existing config or code referencing sliding_context_window will break.
- !Delegation is now disabled by default; crews that relied on agents delegating tasks without explicit configuration will stop delegating on upgrade.
v0.55.2 Sep 13, 2024 · issue -374

CrewAI v0.55.2 adds auto-complete, enriched TaskOutput fields, new install/deploy CLIs, and Pipeline cleanup.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.55.2 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.55.2
```
└──▷ TRY IT
Set up a new CrewAI project and its dependencies in one step.
```
$ crewai install
```
Deploy a CrewAI project to a hosted environment after building it locally.
```
$ crewai deploy
```
Inspect the name and expected output of a completed task result for downstream processing or logging.
python
```
result = crew.kickoff()
for task_output in result.tasks_output:
    print(task_output.name, task_output.expected_output)
```
- ›Adds auto-complete support for CrewAI inputs.
- ›Adds name and expected_output fields to TaskOutput for richer task result inspection.
- ›New crewai install CLI command for streamlined project setup.
- ›New crewai deploy CLI command to deploy CrewAI projects.
- ›Cleans up and stabilizes the Pipeline feature.
v0.51.0 Aug 11, 2024 · issue -375

CrewAI v0.51.0 adds crew testing/evaluation, pipelines, four new tools (Vision, DALL-E, MySQL, NL2SQL), and a sliding context window.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.51.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.51.0
```
└──▷ USE IT
Assign a cheaper or faster LLM specifically to the planning step so your expensive model is reserved for agent execution.
python
```
from crewai import Crew, Agent, Task, Process
from langchain_openai import ChatOpenAI

planner_llm = ChatOpenAI(model="gpt-4o-mini")

crew = Crew(
    agents=[...],
    tasks=[...],
    process=Process.sequential,
    planning=True,
    planning_llm=planner_llm
)
crew.kickoff()
```
Give agents the ability to generate images on demand during a workflow using the new DALL-E Tool.
python
```
from crewai_tools import DallETool
from crewai import Agent

image_agent = Agent(
    role="Image Creator",
    goal="Generate visuals from descriptions",
    backstory="You create images for marketing campaigns.",
    tools=[DallETool()]
)
```
Enable natural-language database querying so agents can answer data questions without writing SQL manually.
python
```
from crewai_tools import NL2SQLTool
from crewai import Agent

db_agent = Agent(
    role="Data Analyst",
    goal="Answer business questions from the database",
    backstory="You query databases using plain English.",
    tools=[NL2SQLTool(db_uri="mysql+pymysql://user:pass@host/dbname")]
)
```
- ›Adds crew testing and evaluation framework for assessing crew performance.
- ›Introduces Pipeline structure for composing multi-crew workflows.
- ›Adds sliding context window to manage long-running agent context automatically.
- ›Supports setting a dedicated LLM for the planning step, separate from agent LLMs.
- ›New crew run CLI command for executing CrewAI projects.
+8 moreshow less
- ›Allows all agent/task attributes to be defined in YAML project configuration.
- ›New Vision Tool for enabling agents to process and reason about images.
- ›New DALL-E Tool for generating images from within agent workflows.
- ›New MySQL Tool for querying MySQL databases from agents.
- ›New NL2SQL Tool for translating natural language queries into SQL.
- ›File-saving now serializes dict outputs as JSON when writing to disk.
- ›Enables verbose settings for tool outputs to aid in debugging and observability.
- ›Adds new GitHub project templates for bootstrapping CrewAI projects.
v0.41.0 Jul 19, 2024 · issue -376

CrewAI v0.41.0 adds crew planning, task replay, memory reset, LLM retry, and type-safe outputs.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.41.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.41.0
```
└──▷ USE IT
Enable pre-execution planning so your crew deliberates a strategy before assigning tasks — useful for complex, multi-step investigations.
python
```
from crewai import Crew

crew = Crew(
    agents=[...],
    tasks=[...],
    planning=True
)
result = crew.kickoff()
```
Replay from a specific failed task after a partial run, without restarting the entire crew from scratch.
```
$ crewai replay <task_id>
```
Clear stale crew memory before re-running to ensure the crew starts fresh without prior context contaminating results.
```
$ crewai reset-memory
```
- ›Adds planning=True to Crew instances so crews reason through a plan before executing tasks.
- ›Introduces a CLI replay feature to list tasks from the last run and re-execute from a specific task.
- ›Enables resetting crew memory before a run via a new reset-memory capability.
- ›Adds LLM call retry support so a failed LLM call no longer halts crew execution.
- ›All crews and tasks now return typed CrewOutput and TaskOutput objects instead of raw strings.
+3 moreshow less
- ›Adds ability to customize the output converter on agents/tasks.
- ›Enhances tools with type hinting and new attributes.
- ›Adds MultiON Tool integration.
└──▷ BREAKING ON UPGRADE
- !All crews and tasks now return TaskOutput and CrewOutput objects instead of plain strings — code that treats crew/task return values as strings will break.
v0.36.0 Jul 6, 2024 · issue -376

CrewAI v0.36.0 adds AgentOps native support, Firecrawl Tools, and tool-result-as-agent-result capability
└──▷ GET THIS VERSION
```
$ git clone --branch v0.36.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.36.0
```
- ›Adds native AgentOps integration for agent observability and monitoring.
- ›Adds Firecrawl Tools for web scraping and crawling within CrewAI agents.
- ›Adds the ability to return a tool's result directly as an agent result.
- ›Improves the coding Interpreter tool.
- ›Adds the ability to create a custom converter class.
v0.35.0 Jun 29, 2024 · issue -377

CrewAI v0.35.0 adds code execution for agents, third-party agent integration, and a new crewai train CLI command.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.35.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.35.0
```
└──▷ TRY IT
Train your crew for a fixed number of iterations before a production run to improve output consistency.
```
$ crewai train -n 5
```
- ›New crewai train -n <X> CLI command lets you train a crew for a specified number of iterations before execution to produce more consistent results.
- ›Agents can now execute code directly as part of a crew workflow.
- ›Supports integrating third-party agents — including LlamaIndex, LangChain, and Autogen agents — as first-class crew members.
v0.32.0 Jun 21, 2024 · issue -377

CrewAI v0.32.0 adds async and per-item kickoff methods, LlamaIndex hub support, and usage metrics on crew output.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.32.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.32.0
```
└──▷ USE IT
Run the same crew independently for each item in a list — useful for processing multiple targets in parallel without managing async yourself.
python
```
results = my_crew.kickoff_for_each(inputs=[{"target": "host1"}, {"target": "host2"}, {"target": "host3"}])
```
Kick off a long-running crew without blocking — lets you launch a crew and continue other work while it runs.
python
```
import asyncio

async def main():
    result = await my_crew.kickoff_async(inputs={"target": "host1"})
    print(result)

asyncio.run(main())
```
Inspect token consumption after a crew run to track LLM costs per execution.
python
```
result = my_crew.kickoff(inputs={"target": "host1"})
print(result.usage_metrics)
```
- ›Adds kickoff_for_each, kickoff_async, and kickoff_for_each_async methods to the crew kickoff API for parallel and per-item execution control.
- ›Adds usage_metrics field to full crew output, exposing token and resource consumption data.
- ›Adds support for all LlamaIndex hub integrations as crew tools.
- ›Adds support for multiple crews in the new YAML format.
- ›Changes the default LLM model to gpt-4o.
+1 moreshow less
- ›Adds timestamps to log output.
└──▷ BREAKING ON UPGRADE
- !The default model is now gpt-4o; crews that relied on the previous default model will switch automatically on upgrade.
v0.30.4 May 13, 2024 · issue -378

CrewAI v0.30.4 adds manager agent override, prompt/response templates for OSS models, and Browserbase and Exa Search tools.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.30.4 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.30.4
```
- ›Adds ability to designate a specific agent as crew manager instead of having the crew auto-generate one.
- ›Adds system, prompt, and response templates so practitioners can tune LLM interaction for open-source and smaller models.
- ›Adds initial support for bringing your own prompts to override built-in crew prompts.
- ›Adds two new built-in tools: Browserbase and Exa Search.
- ›Improves JSON and Pydantic output handling for better compatibility with smaller models.
+2 moreshow less
- ›Improves tool name recognition for better compatibility with smaller models.
- ›Adds ability to automatically create a directory when saving output as a file.
└──▷ BREAKING ON UPGRADE
- !Dependencies have been updated — verify your tool integrations after upgrading.
v0.27.0 Apr 4, 2024 · issue -379

CrewAI v0.27.0 adds shared crew memory, native human input, universal RAG tool support, and custom cache control.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.27.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.27.0
```
└──▷ USE IT
Enable shared crew memory so agents retain context across tasks, improving consistency in multi-step workflows.
python
```
crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    memory=True
)
```
Use a custom cache function on a tool to control exactly when results are cached, e.g. skipping cache for volatile data.
python
```
from crewai_tools import tool

@tool
def my_tool(query: str) -> str:
    ...

my_tool.cache_function = lambda args, result: 'volatile' not in args['query']
```
- ›Adds memory=True parameter to crew configuration to enable shared crew memory, improving outcome reliability (disabled by default).
- ›Adds cache_function attribute to tools for custom caching logic per tool invocation.
- ›Adds native human input support, allowing agents to pause and request input from a human during execution.
- ›Extends RAG tools support to any embedding model and provider, no longer limited to OpenAI.
- ›Adds cross-agent delegation, enabling smoother cooperation and task handoff between agents.
v0.22.0 Mar 11, 2024 · issue -380

CrewAI v0.22.0 adds a crewai create CLI command and dictionary-based agent/task definitions.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.22.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.22.0
```
└──▷ TRY IT
Bootstrap a new CrewAI project scaffold from the command line.
```
$ crewai create
```
Define an agent using a dictionary instead of a class instance, useful for dynamic or config-driven crew setups.
python
```
from crewai import Agent, Task, Crew

agent = Agent(**{
    "role": "Researcher",
    "goal": "Find the latest AI news",
    "backstory": "You are an expert at finding information."
})

task = Task(**{
    "description": "Search for the top 5 AI news stories today",
    "agent": agent
})
```
- ›Adds crewai create CLI command for bootstrapping new CrewAI projects from the command line.
- ›Enables agents and tasks to be defined using Python dictionaries as an alternative to class-based definitions.
- ›Adds clearer agent logging output to improve observability during crew execution.
v0.19.0 Mar 4, 2024 · issue -380

CrewAI v0.19.0 adds execution metrics, input passing at kickoff, and function-calling LLM fallback for agents and crews.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.19.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.19.0
```
└──▷ USE IT
After a crew run, inspect tool usage metrics to audit how many tools were called and where errors occurred.
python
```
result = crew.kickoff(inputs={'domain': 'example.com'})
print(crew.usage_metrics)
```
Fall back to function-calling mode on an agent when the default tool-use strategy is unreliable with your LLM.
python
```
agent = Agent(
    role='Analyst',
    llm=my_llm,
    function_calling_llm=my_function_llm,
    tools=[my_tool]
)
```
- ›Adds function_calling_llm parameter on Agent or Crew to fall back to function calling when standard tool usage fails.
- ›Adds crew.usage_metrics to retrieve execution metrics after a kickoff call, exposing tool usage statistics.
- ›Adds inputs parameter to crew.kickoff(inputs={'key': 'value'}) so runtime values can be injected at execution time.
- ›Enhances delegation capabilities for agents within a crew.
v0.16.0 Feb 28, 2024 · issue -381

CrewAI v0.16.0 adds inputs interpolation and telemetry for tools usage, errors, and token tracking.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.16.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.16.0
```
- ›Adds initial support for inputs interpolation, enabling dynamic input substitution in crew workflows.
- ›Adds ability to track tools usage, tools errors, formatting errors, and token usage within crew runs.
└──▷ BREAKING ON UPGRADE
- !The crewai_tools dependency has been removed; any setup relying on it will break on upgrade.
v0.14.0rc0 Feb 20, 2024 · issue -381

CrewAI v0.14.0rc0 adds crewai-tools integration, Pydantic/JSON task output formatting, and file output support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.0rc0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.14.0rc0
```
- ›Adds support for crewai-tools integration.
- ›Adds support for formatting task output as Pydantic objects or JSON.
- ›Adds support for saving task output to a file.
- ›Supports tools with no arguments.
- ›Revamps tools usage logic to properly use function calling.
+1 moreshow less
- ›Improves reliability for inter-agent delegation.
v0.10.0 Feb 10, 2024 · issue -381

CrewAI v0.10.0 adds full task output capture, step callbacks, multi-argument tool support via JSON, and shared caching across agents.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.10.0
```
└──▷ USE IT
Stream every agent reasoning step in real time — useful for debugging long-running autonomous crews.
python
```
def my_callback(step):
    print(step)

agent = Agent(..., step_callback=my_callback)
crew = Crew(agents=[agent], ..., step_callback=my_callback)
```
- ›Adds full_output return from crew kickoff, exposing all individual task outputs in a single result.
- ›Adds step_callback parameter for both Agents and Crews to receive all intermediate reasoning steps during execution.
- ›New tool usage internals now use JSON, unlocking support for tools that require multiple arguments.
- ›Rebuilt caching structure so multiple agents can share the same cache across a crew run.
- ›Adds opt-in sharing of complete crew run data with the crewAI team.
└──▷ BREAKING ON UPGRADE
- !Removes CrewAgentOutputParser — any code that imports or references it directly will break on upgrade.
v0.5.0 Feb 4, 2024 · issue -381

CrewAI v0.5.0 adds task callbacks, hierarchical process support, task references, and parallel task execution.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.5.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.5.0
```
- ›Adds task callbacks, enabling post-execution hooks on individual tasks.
- ›Adds support for a hierarchical process model for structuring agent workflows.
- ›Adds the ability to reference specific tasks from within another task.
- ›Adds parallel task execution, allowing multiple tasks to run concurrently.
v0.1.32 Jan 14, 2024 · issue -382

CrewAI v0.1.32 adds per-agent iteration limits, RPM throttling, and initial i18n support
└──▷ GET THIS VERSION
```
$ git clone --branch v0.1.32 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.1.32
```
- ›Adds ability to limit the maximum number of iterations for an agent, preventing runaway agent loops.
- ›Adds Request Per Minute (RPM) throttling configurable for both individual Agents and Crews.
- ›Adds initial internationalization (i18n) support with a Greek translation included.
v0.1.14 Dec 30, 2023 · issue -383

CrewAI v0.1.14 adds tool caching, loop execution prevention, verbose logging levels, and pydantic v2 support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.1.14 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.1.14
```
- ›Adds verbose levels to the logger for finer-grained output control.
- ›Adds tool caching to avoid redundant tool calls during agent execution.
- ›Adds loop execution prevention to stop agents from cycling indefinitely.
- ›Expands delegation guidelines to give agents more precise rules for task hand-off.
- ›Updates support to pydantic v2.
└──▷ BREAKING ON UPGRADE
- !Upgrades to pydantic v2, which may break existing model definitions that rely on pydantic v1 behaviour.
v0.1.1 Nov 19, 2023 · issue -384

CrewAI v0.1.1 adds verbose mode for inspecting task execution in real time.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.1.1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.1.1
```
- ›Adds Crew verbose mode to inspect tasks as they are being executed.
v0.1.0 Nov 14, 2023 · issue -384

CrewAI v0.1.0 debuts a Python framework for orchestrating collaborative, role-based autonomous AI agents.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.1.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout v0.1.0
```
└──▷ USE IT
Stand up a minimal crew where two specialized agents tackle a research-then-write task sequentially.
python
```
from crewai import Agent, Task, Crew, Process

researcher = Agent(
    role='Researcher',
    goal='Find key facts about quantum computing',
    tools=[search_tool]
)

writer = Agent(
    role='Writer',
    goal='Summarize research into a short briefing',
    tools=[]
)

research_task = Task(description='Research recent quantum computing breakthroughs', tools=[search_tool])
write_task    = Task(description='Write a 200-word executive summary', tools=[])

crew = Crew(
    agents=[researcher, writer],
    tasks=[research_task, write_task],
    process=Process.sequential
)

result = crew.kickoff()
print(result)
```
- ›Introduces the Agent class for defining autonomous agents with specific roles, goals, and assigned tools.
- ›Introduces the Task class for dynamically creating and assigning tasks with per-task tool specifications.
- ›Introduces the Crew class for grouping agents and coordinating collaborative workflows.
- ›Introduces the Process class supporting sequential task execution for organized, predictable agent pipelines.
- ›Enables inter-agent delegation, allowing agents to autonomously redistribute subtasks among team members at runtime.

CrewAI

Summary

What CrewAI answers

Features

release 10 total

release 7 total

example

release 7 total

release

example

release

release

release

release

release

release

release

release

release

release

release

command line

example

release

example

release

release

release

release 7 total

example

release

release

release

release

example

release

release

release

release

release

Examples

Command line

Release history