bumblehive MCP server
Commit Nov 6, 2026Build complete agent loops using a lightweight Python SDK that manages model calls, tool integration, and resource lifecycles.
Also known as wxhcore/bumblehive
- Stars
- 9
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 6, 2026
- License
- Apache-2.0
What is bumblehive MCP?
bumblehive builds complete agent loops using a lightweight Python SDK that manages model calls, tool integration, and resource lifecycles.
Summary generated from the project's README
About bumblehive MCP server
Developers building autonomous agent loops can use this lightweight Python SDK to manage model calls, tool integration, and resource lifecycles. It provides a structured way to implement agentic behavior by handling the complexities of the agent loop through a central Runtime. Users can choose between running agents in a terminal via run_console, obtaining structured results for applications using run, or consuming runtime events through stream.
Setting up the environment requires Python 3.11 or higher and specific environment variables including BUMBLEHIVE_MODEL, BUMBLEHIVE_API_KEY, and BUMBLEHIVE_BASE_URL. The SDK allows for extending agent capabilities by adding Python functions, built-in tools, or connecting to external MCP servers. It also supports advanced features like conversation history, persisted sessions, and Skills for loading task instructions.
What you can do with it
- 1Debug agents in the terminalA developer can quickly test an agent's response to a prompt using the console method.
- 2Integrate agents into Python appsYou can use the SDK to get structured data back from an agent to use within a larger software project.
- 3Build custom agent interfacesDevelopers can create unique user interfaces by listening to real-time runtime events.
- 4Extend agent functionality with toolsYou can provide your agent with specific capabilities by registering Python functions or MCP servers.
Written from the project's README. Every use case is checked against a quote from it.
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablewxhcore/bumblehive is public and reachablegithub.com/wxhcore/bumblehive ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-06github.com/wxhcore/bumblehive ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHub
小核心,大轰鸣 | Small Core, Big Buzz
A lightweight Python SDK for building a complete Agent Loop in just a few lines of code.
English | 简体中文
Try It in 30 Seconds
Install the Python SDK:
python -m pip install bumblehive
Set BUMBLEHIVE_MODEL, BUMBLEHIVE_API_KEY, and BUMBLEHIVE_BASE_URL, then run:
import asyncio
import os
import bumblehive
async def main():
config = bumblehive.RuntimeArguments(
model=os.environ["BUMBLEHIVE_MODEL"],
api_key=os.environ["BUMBLEHIVE_API_KEY"],
base_url=os.environ["BUMBLEHIVE_BASE_URL"],
skill_names=[],
tool_names=[],
)
async with bumblehive.from_config(config) as runtime:
await runtime.run_console("Explain Agent Runtime in one sentence.")
asyncio.run(main())
run_console() displays the answer directly in your terminal. The empty lists disable Skills and tools for this first call. See the quickstart for installation and environment setup.
Choose a Run Method
| API | Best for |
|---|---|
run_console() |
Trying or debugging an Agent in the terminal |
run() |
Getting a structured result for a Python application |
stream() |
Consuming runtime events in a custom interface |
Core Capabilities
- Manage model calls, the Agent Loop, and resource lifecycles through Runtime.
- Give an Agent Python functions, built-in tools, or MCP servers.
- Add streaming events, conversation history, persisted sessions, Skills, and observability hooks when needed.
Next Steps
The English documentation is in progress. See the English overview, the complete Chinese documentation, or the runnable examples.
| Goal | Start here |
|---|---|
| Make the first Agent call | Quickstart (Chinese) |
| Add a Python tool | Register a tool (Chinese) |
| Save a conversation | History and sessions (Chinese) |
| Load task instructions | Skills (Chinese) |
| Connect external tools | MCP (Chinese) |
| Look up a public API | API Reference (Chinese) |
| Browse runnable code | Examples |
Local Development
The Python SDK, Server, and WebUI development workflow supports macOS, Windows, and Ubuntu Linux. It requires Python 3.11+, Node.js 22.12+, and pnpm 10.33.0. A dedicated Conda environment is recommended:
pnpm run setup
pnpm run setup is the only project setup entry point. It installs every Node workspace dependency from the root lockfile, then uses the active Python interpreter to install the SDK, Server, test, and desktop packaging dependencies before running the core environment check. pnpm run dev repeats that lightweight preflight before starting either process. If a requirement is missing, the command stops with a targeted error. The same command works on Ubuntu Linux and does not require the desktop system toolchain. Set BUMBLEHIVE_PYTHON to an interpreter path to override the active interpreter.
For daily development:
pnpm run dev
This starts the Server on 127.0.0.1:18421 and the WebUI on 127.0.0.1:1420. Use pnpm run dev:server or pnpm run dev:web to run one component, and pnpm test for the Python test suite.
Desktop
The optional desktop workflow currently targets macOS and Windows. macOS additionally requires Rust and Xcode Command Line Tools. Windows additionally requires the Rust MSVC toolchain, WebView2, and Microsoft C++ Build Tools with the Desktop development with C++ workload.
After the shared pnpm run setup, start development or create the installer for the current platform:
pnpm run dev:desktop
pnpm run build:desktop
Both commands verify Rust, Tauri, and the platform toolchain, then automatically package the Python Server as a sidecar before starting Tauri. build:desktop creates an app and DMG on macOS or an NSIS installer on Windows.
bumblehive MCP questions
Can I connect external tools to the agent?
Yes, you can provide agents with Python functions, built-in tools, or MCP servers.