Skip to main content

crewai-lumify

CrewAI integration for Lumify Sports Intelligence — the agent-ready sports intelligence API. Connect Lumify's hosted MCP server (schedules, live scores, odds, public betting splits, and explainable AI bet confidence across 8+ sports) to any CrewAI agent in one call.

It's a thin, typed helper that returns a ready-to-use MCPServerHTTP for CrewAI's Agent(mcps=[...]) DSL, pointed at https://lumify.ai/mcp.

Install

pip install crewai-lumify
# Agent examples also need a model provider, e.g.:
# pip install 'crewai[openai]'

Or install the latest from source:

pip install "git+https://github.com/lumifyai/lumify.git#subdirectory=clients/crewai-lumify"

Get a key

Grab a free instant key in seconds — no signup, email, or card at https://lumify.ai/docs/ai (100 credits, 14-day expiry). For a persistent account with 1,000 starter credits, use https://lumify.ai/api-keys.

Set it as LUMIFY_API_KEY, or pass it explicitly. get_lumify_mcp requires a key by default so the first tool call does not fail with an opaque 401. Pass require_api_key=False only if you want to introspect the public tools/list catalog. Raw lmfy-... values and Bearer lmfy-... copy-pastes are both accepted.

Quick start

from crewai import Agent, Task, Crew
from crewai_lumify import get_lumify_mcp

agent = Agent(
    role="Sports Analyst",
    goal="Find edges using odds and intelligence",
    backstory="Expert with access to sports market data",
    mcps=[get_lumify_mcp()],  # reads LUMIFY_API_KEY
)

task = Task(
    description="What's the best MLB bet today, with the rationale?",
    expected_output="A short recommendation with confidence and why.",
    agent=agent,
)

result = Crew(agents=[agent], tasks=[task]).kickoff()
print(result)

Pass a key explicitly or point at a self-hosted endpoint:

mcp = get_lumify_mcp(api_key="lmfy-...", url="https://lumify.ai/mcp")
agent = Agent(..., mcps=[mcp])

Filter to a subset of tools with CrewAI's filter helpers:

from crewai.mcp.filters import create_static_tool_filter
from crewai_lumify import get_lumify_mcp

mcp = get_lumify_mcp(
    tool_filter=create_static_tool_filter(
        allowed_tool_names=["list_events", "get_odds", "get_intelligence"],
    ),
)

Security note: CrewAI's MCPServerHTTP stores your Authorization header verbatim, and its default repr() includes it. Avoid print(mcp), print(agent), or verbose/debug logging that dumps agent or MCP server config — it can leak your Lumify API key into logs or crash reports.

Tools

Loads the full Lumify MCP tool surface (18 tools): list_sports, list_seasons, list_events, get_event, batch_get_events, query_events, get_live_score, get_odds, get_odds_history, get_stats, get_splits, get_intelligence, list_teams, get_team, search_players, get_player, get_player_events, and estimate_cost (free pre-call credit-cost estimate).

Links

License

MIT

Release files for crewai-lumify 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for crewai-lumify 0.1.0
File Size Uploaded
crewai_lumify-0.1.0.tar.gz 25.1 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for crewai-lumify 0.1.0
File Interpreter ABI Platform
crewai_lumify-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 31.5 kB

Release files / crewai_lumify-0.1.0.tar.gz

Download URL crewai_lumify-0.1.0.tar.gz
Size 25.1 kB
Tags Source
SHA-256 checksum
How to use checksums
ab36aa1ef9a9e3a733b86fecc35004caad6eaa09e208cc543f163b4d8b19d36d
BLAKE2b-256 checksum
How to use checksums
f46a53a923b9ae6b32636556e41ba821c8d8de9b2ace2459412326850a84f610
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.12

Release files / crewai_lumify-0.1.0-py3-none-any.whl

Download URL crewai_lumify-0.1.0-py3-none-any.whl
Size 6.4 kB
Tags Python 3
SHA-256 checksum
How to use checksums
9fcb0ff074d08083224a2a07fa1c760acce39a4b69cd48c4b1be9ed35266efdd
BLAKE2b-256 checksum
How to use checksums
45af84e70891284ae5ac55c4d09711b546a75d4477e2c320ec1cd8e09d9186a4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.12

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page