Zero-dependency codebase bundler with automatic secret redaction and token budgeting. Includes an stdio Model Context Protocol (MCP) server for local LLM tools.
Lightning-fast, AST-aware repository bundler & live MCP context server for AI pair programming.
repoflux solves the single biggest bottleneck in LLM-assisted software engineering: context overhead and safety.
Feeding messy codebases into ChatGPT, Codex, Claude, or Cursor causes truncation, prompt inflation, and catastrophic API key leaks. repoflux analyzes your repository in milliseconds, maps token weights, scrubs leaked credentials, and compiles a clean, token-budgeted representation.
It also doubles as a native Model Context Protocol (MCP) server, giving Claude Desktop, Cursor, and agentic IDEs direct live access to your codebase.
- 🏎️ Zero-Dependency Core: Pure TypeScript engine running at sub-60ms speeds across multi-thousand file repositories. - 🔒 Automated Secret Redaction: Automatically detects and redacts OpenAI, Anthropic, AWS, Slack, GitHub, Stripe tokens, JWTs, and private keys before LLMs see them. - 📊 Token-Aware Tree: Visualizes file weights and distribution so you can trim bloated files before pasting into ChatGPT/Codex. - 🎯 Token Budgeting: Specify --max-tokens to guarantee your bundle never exceeds model context windows (e.g., 32k, 128k, 200k). - 🧩 Native MCP Server: Run repoflux --mcp to expose getrepomap, packcodebase, and auditsecrets directly to Claude Desktop & Cursor. - 🛡️ Intelligent .gitignore: Automatically respects .gitignore rules and rejects binary files, lockfiles, nodemodules, and build outputs.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/dirka111q-maker--repoflux.html)
ChatGPT thinks. Codex works. Use ChatGPT as the planning brain while keeping the Codex harness.
shinthink/blitzstrike⚡ Blitz Strike — a universal MCP penetration-testing toolbelt. Structured methodology: reconnaissance & attack-surface mapping, source-to-sink analysis, and live validation. 57 escalation chains, 130-
0xwilliamortiz/openclaude-improvedruns anywhere. uses anything
halofyai/halofyHalofy is the open access and governance layer for AI agents across your organization. Identity, policy, provenance, audit, and signed erasure.
CaptureGrubEnchant/SolidWorksSolidWorks MCP Server connects an AI assistant to a running SolidWorks instance. Sketch, extrude, fillet, export STEP/STL, generate macros.
arielshad/3d-asset-server3d assets api and mcp
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.