tenets

Intelligent Context Ranking

Multi-factor ranking using BM25, TF-IDF, AST analysis, import graphs, and Git history to find the most relevant code for any task

MCP Server Integration

Native integration with Cursor, Claude Code, Windsurf, and VS Code via Model Context Protocol for seamless AI assistant enhancement

Guiding Principles (Tenets)

Define persistent rules that shape how AI understands your codebase — coding standards, architectural decisions, and team conventions

Code Quality Metrics

Cyclomatic complexity, cognitive complexity, Halstead metrics, maintainability indexing, and technical debt tracking

100% Local Processing

Everything runs on your machine. No cloud APIs, no data uploads, no telemetry. Your code stays completely private

Session Persistence

SQLite-backed sessions that remember context across multiple AI interactions. Branch sessions to explore alternatives

Dependency Visualization

Map and visualize import relationships, dependency trees, and module coupling to understand codebase architecture

Team Productivity Analytics

Velocity tracking, hotspot analysis, contributor analytics, and development intelligence for engineering managers

MCP server focused on providing up-to-date library documentation for AI assistants

AI coding assistant with built-in codebase context and Git integration

Standardized protocol for sharing codebase context with AI tools

Enterprise code intelligence with AI assistant and codebase search

Tenets is exactly what the AI coding ecosystem needs: an intelligent, privacy-first context engine that makes every AI assistant smarter about your specific codebase. The multi-factor ranking (BM25, TF-IDF, import graphs, Git history) produces genuinely better context than simple file inclusion. The guiding principles feature is uniquely powerful — it lets you encode team knowledge and architectural decisions into persistent rules that shape every AI interaction. Being free and open-source removes any barrier to trying it. The only friction is the Python/pip requirement and initial setup. For any developer using AI coding assistants, Tenets is a must-install. Rating: 4.5/5

Is Tenets free?

Yes, Tenets is completely free and open-source under the MIT license. There are no paid tiers, no usage limits, and no plans to add paid features. It will always be free.

Does Tenets send my code to the cloud?

No. Tenets runs 100% locally on your machine. No cloud APIs, no data uploads, no telemetry. Your codebase stays completely private.

What AI assistants does Tenets work with?

Tenets integrates with Cursor, Claude Code (Claude Desktop), Windsurf, and VS Code via the Model Context Protocol (MCP). It also provides a CLI and Python library for custom integrations.

What are 'tenets' in Tenets?

Tenets (guiding principles) are persistent rules you define that shape how AI interprets your codebase. They encode coding standards, architectural decisions, and team conventions so AI assistants follow your team's practices.

How does Tenets rank code relevance?

Tenets uses a multi-factor pipeline: BM25 and TF-IDF for text relevance, AST analysis for code structure, import graph analysis for dependency relationships, and Git history for change patterns. This produces higher-quality context than simple keyword matching.

Does Tenets slow down my IDE?

Tenets runs as a background MCP server and is designed to be lightweight. Initial indexing may take a few minutes on large codebases, but subsequent queries are fast.

Can I use Tenets with Python projects?

Yes, Tenets works with any programming language. It analyzes code structure, imports, and Git history regardless of the language. Python projects get additional benefits from the native AST analysis.

How do I install Tenets?

Install with pip: 'pip install tenets[mcp]' for MCP server functionality, or 'pip install tenets[all]' for all features including ML and visualization. Then configure it as an MCP server in your IDE.