Universal AI context generator that saves thousands of tokens per conversation in Claude Code, Cursor, Copilot, and more.
codesight is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
Repository setup guidance
gh repo view Houseofmvps/codesight --webOpen the official repository or website.
Check the README for package manager, auth, and platform requirements.
Try it in a small test task inside your agent workflow.
Codesight analyzes your codebase and generates a compact context file for AI coding assistants. It uses AST precision for TypeScript and regex for other languages, detecting frameworks, routes, schemas, and dependencies. This reduces token usage and helps AI tools understand your project faster.
Codesight is a zero-dependency CLI tool that generates optimized context for AI coding assistants like Claude Code, Cursor, GitHub Copilot, and OpenAI Codex. It scans your project's source code, identifies frameworks, routes, database schemas, components, and dependencies, then produces a concise markdown summary. This summary can be used as a system prompt or attached to conversations, saving thousands of tokens per session. Codesight supports 14 programming languages and over 30 frameworks. It includes features like wiki knowledge base generation, blast radius analysis, MCP server mode with 13 tools, and benchmark mode to show token savings. It is designed to be run with a single npx command, requiring no configuration or API keys.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 70 days ago.
1104 GitHub stars indicate community interest.
0 open issues signal maintenance load.
MIT license detected.
Reduce token usage in AI coding assistant conversations by providing a concise project overview.
Quickly onboard AI tools to a new or unfamiliar codebase by generating a structured context file.
Analyze the impact of changes with blast radius detection before refactoring.
Generate a persistent wiki knowledge base for consistent AI context across sessions.
Optimize AI tool configuration for specific assistants like Claude Code or Cursor.
Codesight reads your entire project directory; ensure you run it only on trusted codebases.
Generated context files may contain sensitive information if your code includes secrets or credentials.
1,104
Stars
102
Forks
0
Issues
MIT
License
One command gives AI agents instant codebase context, replacing 50,000+ tokens of exploration with ~250 tokens.
A fast Rust tool to serialize text files in a repo for LLM consumption.
A memory layer for coding agents that stores and retrieves repo context, conventions, and invariants.
Universal AI context generator that saves thousands of tokens per conversation in Claude Code, Cursor, Copilot, and more.
Developers using AI coding assistants who want to reduce token costs and improve context quality., Team leads and architects managing large codebases who need to onboard AI tools efficiently., Open source maintainers who want to provide AI-friendly documentation for their projects.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Codesight reads your entire project directory; ensure you run it only on trusted codebases. Generated context files may contain sensitive information if your code includes secrets or credentials.
Similar or complementary options to evaluate include Stacklit, yek, gps.
Jun 6, 2026
2 security/trust notes recorded.
Setup difficulty is 1/5.