syncable-dev/memtrace-public

Structural memory for AI coding agents. Bi-temporal graph, MCP-native, zero LLM calls. Cursor · Claude Code · Codex · Hermes · VS Code · Windsurf.

What it solves

Memtrace provides "structural memory" for AI coding agents, preventing them from losing context or breaking dependencies when working on large codebases. It eliminates the need for agents to re-read files constantly and solves the problem of "blind refactors" where agents make changes without understanding the full blast radius of their edits.

How it works

Memtrace transforms a codebase into a live, bi-temporal episodic structural knowledge graph. Unlike general memory systems that use LLMs to infer relationships, Memtrace uses Rust and Tree-sitter AST parsers to deterministically resolve symbols (functions, classes, interfaces) and their relationships (calls, imports, implementations) locally.

It adds two critical dimensions to standard AST graphs:

  1. Temporal Memory: Tracks the full version history of every symbol, allowing agents to query what changed, what is unexpected, or what will break based on six different scoring algorithms.
  2. Cross-service API Topology: Maps HTTP call graphs between different repositories to detect service-to-service dependencies.

Who it’s for

Developers using AI coding agents (such as Claude Code, Cursor, Windsurf, or VS Code Copilot) who need their agents to have a deep, persistent understanding of complex, multi-language repositories.

Highlights

  • High Performance: Indexes 1,500 files in under 2 seconds with zero API costs, significantly faster than LLM-based memory systems.
  • MCP Native: Exposes over 25 tools via the Model Context Protocol for search, relationship analysis, impact analysis, and graph algorithms.
  • Broad Language Support: Supports 20+ languages and framework-aware scanners for backend HTTP (Express, FastAPI, Django, etc.), CI/CD (GitHub Actions), and databases (PostgreSQL RLS).
  • Privacy-First: Runs entirely on the local machine; source code never leaves the environment.
  • Agent Skills: Includes 17 pre-built skills/guidance prompts that teach agents how to chain tools for tasks like incident investigation and refactoring guides.