# Houseofmvps/codesight

Universal AI context generator. Saves thousands of tokens per conversation in Claude Code, Cursor, Copilot, Codex, and more.

Repository: https://github.com/Houseofmvps/codesight
Canonical: https://ross.abutalabs.com/products/codesight
Homepage: https://github.com/Houseofmvps/codesight
Language: TypeScript
License: MIT
License Family: permissive
Topics: ai, claude, cli, code-analysis, codebase, codex, context-engineering, copilot, cursor, developer-tools, llm, mcp, repo-map, token-savings
Last push: 2026-07-27T13:30:50+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 94, release rhythm 35, longevity 10
- inputs: {"age_days": 151, "days_push": 37, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1385, forks 124 (observed 2026-08-28T04:04:34.859898+00:00)

## What it is
codesight is a zero-dependency Node.js CLI that scans a codebase and generates a compact AI context map (routes, schemas, components, dependencies) to save tokens in AI coding assistants. It also ships 14 MCP tools and can emit CLAUDE.md, .cursorrules, and other assistant config files.

## Use cases
- generate a codebase context map for Claude Code or Cursor
- save tokens when using AI coding assistants on large projects
- auto-generate CLAUDE.md, .cursorrules, and AGENTS.md files
- map routes, schemas, and components across a multi-language repo
- run an MCP server exposing codebase analysis tools
- produce a wiki knowledge base from a repository

## When to choose
- you use AI coding tools and want to cut context/token costs
- you need a quick, config-free repo overview across 14 languages
- you want assistant rule files generated automatically
- you want MCP tools for codebase introspection

## When to avoid
- you need deep semantic analysis beyond regex detection for non-TypeScript languages
- you need a fully offline LLM or code search engine rather than a context generator
- your project is in a language outside its supported list

## Facets
- artifact type: cli-tool
- maturity: active
- function: developer-tools, cli, mcp, parser, documentation, llm-inference
- domain: developer-tools, large-language-models, documentation
- platform: cli, cross-platform
- tags: context-engineering, token-savings, repo-map, code-analysis, claude-code, cursor, copilot, codex, mcp-server, zero-dependencies, command-line, ai-agents, nodejs

## Member repositories
- Houseofmvps/codesight (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:34.859898+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:39:55.816634+00:00, confidence not recorded.
  - readme: https://github.com/Houseofmvps/codesight (fetched 2026-08-28T04:04:34.859898+00:00, sha b13c63719278)
  - homepage: https://github.com/Houseofmvps/codesight (fetched 2026-08-29T11:55:53.178084+00:00, sha cff7c7b21fa0)
  - registry_npm: https://registry.npmjs.org/codesight (fetched 2026-08-29T11:55:53.188297+00:00, sha c4cdce589fa9)
- Data as of 2026-08-30T08:39:29.467469+00:00.