CICADA

Provides token-efficient, structured access to Elixir, Python, and Erlang codebases for AI assistants.

typescript

19

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "wende-cicada": {
      "command": "cicada-mcp",
      "args": [],
      "env": {
        "CICADA_CONFIG_DIR": "PLACEHOLDER_PATH"
      }
    }
  }
}

CICADA MCP Server provides fast, token-efficient access to codebases by serving structured, indexed knowledge to AI assistants. It enables precise code discovery, cross-references, and attribution for Elixir, Python, and Erlang projects while keeping your local repository fully private.

How to use

Install and run the MCP server through your editor’s integrated MCP management. Once active, you interact with CICADA through your preferred MCP client (Claude/Cursor/VS/Gemini/CodeX/OpenCode, etc.). Ask your assistant to perform targeted code exploration, such as finding where a function is defined and used, examining API surfaces, or tracing PR-related changes.

How to install

# 1. Install uv (if needed)
# curl -LsSf https://astral.sh/uv/install.sh | sh
uv tool install cicada-mcp

# In your repo
cicada claude   # or: cicada cursor, cicada vs, cicada gemini, cicada codex, cicada opencode

Optional quick try before permanent installation runs CICADA on demand with lower indexing quality but no install required.

uvx cicada-mcp claude   # or cursor, vs

or

gemini mcp add cicada uvx cicada-mcp

codex mcp add cicada uvx cicada-mcp

These commands install CICADA through your editor’s MCP management so you can start working with it in your projects right away.

Configuration and usage notes

After installation, you can manage CICADA from the editor’s MCP interface. Use the following commands to control and inspect the index and integration.

Troubleshooting tips

If you encounter issues connecting to the MCP server, verify that the MCP component is installed and that the local index exists for your project. Restart the editor completely if the MCP session is unresponsive.

Available tools

query

Smart code discovery with keyword vs pattern detection, filters, and concise results that guide the next steps.

search_module

Deep module analysis showing complete API surfaces, signatures, and docs with bidirectional usage context.

search_function

Find function definitions and all call sites with usage context and optional usage examples.

git_history

Unified git history tool for blame, commits, and PR evolution across files and functions.

expand_result

Drill down from a result to auto-expand related modules or functions and reveal dependencies.

find_dead_code

Identify potentially unused functions with confidence levels and provide cleanup guidance.

query_jq

Advanced index queries using jq syntax for power users to extract precise insights.