Tenets

Local MCP server for AI coding assistants that aggregates code context and injects guiding tenets.

python

6

GitHub Stars

python

Language

5 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": {
    "jddunn-tenets": {
      "command": "tenets-mcp",
      "args": []
    }
  }
}

Tenets MCP Server operates locally to gather and rank code context for AI coding assistants, and automatically injects your guiding principles into every prompt. It keeps all processing on your machine, so there are no external API costs or data leaving your environment.

How to use

Install and start the MCP server locally so your AI assistant can fetch and summarize relevant code while respecting your guiding principles.

Use with an MCP client

Start by running the MCP server locally, then connect your AI assistant client to the server for context-aware prompts.

Example client configurations

{
  "mcpServers": {
    "tenets": { "type": "stdio", "command": "tenets-mcp", "args": [] }
  }
}

Starting the MCP server in practice

Prerequisites: Ensure you have Python and a compatible package manager installed. Then install the MCP package and start the server.

pipx install tenets[mcp]
tenets-mcp

Connecting clients by example (manual steps)

Claude Code, Cursor, Windsurf, or VS Code can interface with the MCP server once you provide the stdio configuration. Use the following example in your client settings to connect to the local server.

{ "mcpServers": { "tenets": { "type": "stdio", "command": "tenets-mcp", "args": [] } } }

Available tools

distill

Build Context with content by finding and aggregating relevant files for a given query.

rank

Preview files without content and understand why they are ranked, with optional ML-enhanced ranking.

session_pin_folder

Pin important files to a session to guarantee inclusion in context.

session_create

Create a workspace session to hold your guiding tenets and context.

tenet_add

Add a guiding principle (tenet) with priority levels to influence all prompts.

instill

Inject guiding principles into the current or a named session so every prompt includes them.

distill --ml

Enable ML embeddings for semantic understanding during context build.

examine

Analyze codebase for complexity and other quality metrics.