CICADA provides structured, token-efficient access to code across 17+ programming languages, enabling AI code assistants to perform contextual analysis and discovery. It focuses on reducing unnecessary context and improving semantic understanding of code.
From the registry: Code intelligence MCP server for 17+ programming languages.
$ uv tool install cicada-mcpPlease install the `cicada` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install uv** — Install the uv Python package manager (https://github.com/astral-sh/uv)
- **Install cicada-mcp** — Install cicada-mcp as a uv tool Run: `uv tool install cicada-mcp`
Canonical MCP server config (stdio transport):
- command: `cicada-mcp`
- args: []
- optional environment variables:
- `CICADA_CONFIG_DIR`: Path to the CICADA config directory for this project (example: `~/.cicada/projects/<hash>`)
Note: Code-intelligence MCP for 17+ languages (Elixir, Python, TypeScript, Rust, etc.). Before use, run `cicada <editor>` in project root to index (e.g., `cicada claude`). Pass --watch for auto-reindexing.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.CICADA_CONFIG_DIRrequiredSpecifies the directory for CICADA configuration.AI Agents Framework with Self Reflection and MCP support