ContextSkeleton
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
β‘ ContextSkeleton
Zero-latency Structural Code Folding CLI, Model Context Protocol (MCP) Server & Token Optimizer for AI Agents (Cursor, Claude Code, Windsurf, AGY).
ContextSkeleton folds function and class implementations while preserving top-level signatures, interface definitions, exported types, and docstrings across JavaScript, TypeScript, Python, Go, and C-style languages.
Token reduction scales with function size and implementation depth β typically 0% on trivial 1-2 line utility functions (comment-marker overhead exceeds the savings), 45-70% on production-sized modules with substantial function bodies, and up to 90%+ on large, implementation-heavy files. Across our internal benchmark set spanning small, medium, and large files in TypeScript, Python, Go, and Rust, the blended average was 67%.
π Quick Start
1. Run CLI Scan
Scan your project to view code compression and token savings metrics:
npx context-skeleton scan2. Copy Formatted Context for AI Chat Prompts
Dump the compressed skeleton context directly to stdout for Cursor, Claude Code, or Windsurf:
npx context-skeleton copy > context.md3. Add to PRs & .cursorrules / CLAUDE.md
Inject automated token-savings badges into your repository:
npx context-skeleton badgeRelated MCP server: cctx-mcp
π€ Model Context Protocol (MCP) Integration
Integrate ContextSkeleton natively with Claude Code, Cursor, or AGY agents.
Add to your claude_desktop_config.json or .cursorrules:
{
"mcpServers": {
"context-skeleton": {
"command": "npx",
"args": ["-y", "context-skeleton-mcp"]
}
}
}Exposed MCP Tools:
get_repo_skeleton: Returns folded structural skeleton of target repository.unfold_symbol: Retrieves exact implementation of a specific function or class on demand.get_token_savings: Returns exact token & prompt cost savings metrics.
π Features & Benchmarks
Feature | Raw Codebase | With ContextSkeleton | Benefit |
Blended Benchmark (9 files) | 6,183 tokens | 2,028 tokens | 67.2% Blended Savings |
Production Modules (88-135 lines) | ~550 tokens / file | ~220 tokens / file | 45% to 70% Reduction |
Large Files (500+ lines) | 2,963 tokens | 236 tokens | 92.0% Reduction |
Syntax Errors | Common (unpruned) | 0% (Valid Signatures) | Syntactically intact |
Note: Token reduction scales with function size and implementation depth β typically 0% on trivial 1-2 line utility functions (comment-marker overhead exceeds the savings), 45-70% on production-sized modules with substantial function bodies, and up to 90%+ on large, implementation-heavy files. Across our internal benchmark set spanning small, medium, and large files in TypeScript, Python, Go, and Rust, the blended average was 67%.
π» Tech Stack ($0 Infra Cost)
Engine: Zero-dependency Node.js Structural Signature Pruner & Token Counter
MCP Server: Stdio JSON-RPC 2.0
Web App: Single-Source HTML5 + Modern CSS + Pure JS
π License
MIT License Β© 2026 ContextSkeleton
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
Related MCP Connectors
An MCP server that gives your AI access to the source code and docs of all public github repos
LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.
Repository knowledge graph MCP server for codebase understanding and debugging.
MCP server for progressive tool usage at any scale (see https://klavis.ai)
Related MCP Servers
- AlicenseAqualityBmaintenanceMCP server that reduces token consumption in AI coding assistants by up to 90% via structural reads, PreToolUse hooks, and tp-\* subagents.25280 npm5MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that provides structure-aware code analysis (symbol trees, dependencies, docs) to reduce AI agent token consumption by up to 99%, along with Git commit intelligence.MIT
- AlicenseAqualityDmaintenanceAn AST-based MCP server that provides token-efficient codebase skeletons to LLM agents, reducing context token usage by 80-95% by exposing structural information instead of full source files.53MIT
- AlicenseNot gradedqualityCmaintenanceMCP server that reduces AI agent token usage by up to 90% through intelligent context compression. Enables efficient code exploration, multi-file refactoring, and debugging by providing tools for smart reading, searching, and managing code context.4MIT