jellyfish-mcp
by Jellyfish-AI
The Jellyfish MCP server allows a host to interact with your Jellyfish instance, enabling natural language queries about your engineering metrics, team data, and other information available through the Jellyfish API. It provides a Model Context Protocol server for retrieving and analyzing data from Jellyfish's API.
Last updated: N/A
What is jellyfish-mcp?
The jellyfish-mcp server is a Model Context Protocol (MCP) server designed to facilitate interaction with the Jellyfish API. It allows applications like Claude Desktop or Cursor to query and analyze data from your Jellyfish instance using natural language.
How to use jellyfish-mcp?
To use the server, you need to generate an API token from your Jellyfish instance, clone the repository, install dependencies using uv, configure your credentials (preferably using keyring), and then configure your host application (e.g., VSCode or Claude Desktop) to connect to the server. The server can then be queried using natural language prompts within the host application.
Key features of jellyfish-mcp
Retrieves API schema
Lists available API endpoints
Makes GET requests to Jellyfish API
Provides tools for allocations, delivery, metrics, people, and teams data
Supports PromptGuard 2 for prompt injection attack prevention
Use cases of jellyfish-mcp
Querying engineering metrics using natural language
Retrieving team data and allocations
Analyzing deliverable scope and effort history
Listing engineers and teams
Integrating Jellyfish data with applications like Claude Desktop and VSCode
FAQ from jellyfish-mcp
How do I generate an API token?
How do I generate an API token?
Go to the Data Connections page in your Jellyfish instance, click the API Export tab, and generate a new token.
What is the preferred method for setting the Jellyfish Export API token?
What is the preferred method for setting the Jellyfish Export API token?
The preferred method is to use keyring, which comes with jellyfish-mcp. It is the most secure and ensures credentials are stored using your operating system's preferred credential store.
How do I configure the server with VSCode?
How do I configure the server with VSCode?
Open the Command Palette, search for 'MCP: Add Server...', and provide the path to the uv executable and the server.py file.
How do I configure the server with Claude Desktop?
How do I configure the server with Claude Desktop?
Create or edit your Claude Desktop configuration file and add a mcpServers entry with the path to the uv executable and the server.py file.
How do I enable PromptGuard 2?
How do I enable PromptGuard 2?
Create an account on https://huggingface.co, request access to Llama models, create a huggingface API token, and run uv run llamafirewall configure within jellyfish-mcp.