[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:tokensave":3},"\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Faovestdipaperino\u002Ftokensave\u002FHEAD\u002Fsrc\u002Fresources\u002Flogo.png\" alt=\"TokenSave\" width=\"300\" \u002F>\n\u003C\u002Fp>\u003Ch3>Semantic Code Intelligence for AI Coding Agents\u003C\u002Fh3>\u003Cp align=\"center\">\u003Cstrong>Fewer tokens • Fewer tool calls • 100% local\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fcrates.io\u002Fcrates\u002Ftokensave\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fcrates\u002Fv\u002Ftokensave.svg\" alt=\"crates.io\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg\" alt=\"License: MIT\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.rust-lang.org\u002F\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRust-1.70+-orange.svg\" alt=\"Rust\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FmacOS-supported-blue.svg\" alt=\"macOS\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLinux-supported-blue.svg\" alt=\"Linux\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWindows-supported-blue.svg\" alt=\"Windows\" \u002F>\n  \u003Ca href=\"https:\u002F\u002Fhypercommit.com\u002Ftokensave\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHypercommit-DB2475\" alt=\"Hypercommit\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Chr \u002F>\n\u003Ch2>Why tokensave?\u003C\u002Fh2>\n\u003Cp>AI coding agents waste tokens exploring codebases. Every grep, glob, and file read costs money. On complex tasks, agents spawn multiple Explore sub-agents that scan hundreds of files just to build context.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>tokensave gives agents a pre-indexed semantic knowledge graph.\u003C\u002Fstrong> Instead of scanning files, the agent queries the graph and gets instant, structured answers -- the right symbols, their relationships, and source code, in one call.\u003C\u002Fp>\n\u003Ch3>How It Works\u003C\u002Fh3>\n\u003Cpre>\u003Ccode>┌──────────────────────────────────────────────────────────────┐\n│  AI Coding Agent (Claude Code, Codex, Gemini, Cursor, ...)   │\n│                                                              │\n│  \"Implement user authentication\"                             │\n│        │                                                     │\n│        ▼                                                     │\n│  ┌─────────────────┐       ┌─────────────────┐               │\n│  │  Sub-agent      │ ───── │  Sub-agent      │               │\n│  └────────┬────────┘       └─────────┬───────┘               │\n└───────────┼──────────────────────────┼───────────────────────┘\n            │                          │\n            ▼                          ▼\n┌──────────────────────────────────────────────────────────────┐\n│  tokensave MCP Server                                        │\n│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐           │\n│  │   Search    │  │   Callers   │  │   Context   │           │\n│  │   \"auth\"    │  │  \"login()\"  │  │   for task  │           │\n│  └──────┬──────┘  └──────┬──────┘  └──────┬──────┘           │\n│         └────────────────┼────────────────┘                  │\n│                          ▼                                   │\n│              ┌───────────────────────┐                       │\n│              │   libSQL Graph DB     │                       │\n│              │   • Instant lookups   │                       │\n│              │   • FTS5 search       │                       │\n│              └───────────────────────┘                       │\n└──────────────────────────────────────────────────────────────┘\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>\u003Cstrong>Without tokensave:\u003C\u002Fstrong> Agents use \u003Ccode>grep\u003C\u002Fcode>, \u003Ccode>glob\u003C\u002Fcode>, and \u003Ccode>Read\u003C\u002Fcode> to scan files -- many API calls, high token usage.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>With tokensave:\u003C\u002Fstrong> Agents query the graph via MCP tools -- instant results, local processing, fewer tokens.\u003C\u002Fp>\n\u003Chr \u002F>\n\u003Ch2>Key Features\u003C\u002Fh2>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>\u003C\u002Fth>\n\u003Cth>\u003C\u002Fth>\n\u003Cth>\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>\u003Cstrong>Smart Context Building\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>\u003Cstrong>Semantic Search\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>\u003Cstrong>Impact Analysis\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>One tool call returns everything the agent needs -- entry points, related symbols, and code snippets.\u003C\u002Ftd>\n\u003Ctd>Find code by meaning, not just text. Search for \"authentication\" and find \u003Ccode>login\u003C\u002Fcode>, \u003Ccode>validateToken\u003C\u002Fcode>, \u003Ccode>AuthService\u003C\u002Fcode>.\u003C\u002Ftd>\n\u003Ctd>Know exactly what breaks before you change it. Trace callers, callees, and the full impact radius of any symbol.\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>80+ MCP Tools\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>\u003Cstrong>50+ Languages\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003Ctd>\u003Cstrong>12+ Agent Integrations\u003C\u002Fstrong>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>From call graph traversal to dead code detection, atomic\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n",1784240408269]