memtrace-public

Overview

Memtrace transforms codebases into live knowledge graphs that AI coding agents can query in milliseconds. It maintains structural memory of functions, classes, call edges, and versions across every session, enabling agents to navigate complex codebases without re-reading files or breaking invisible dependencies. The system uses Tree-sitter AST parsing to build deterministic symbol relationships locally, avoiding expensive LLM inference while maintaining high accuracy in code analysis.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• find_code

  Hybrid search using BM25 + semantic + RRF to find code in the repository

  Try:Find functions that handle user authenticationTry:Search for all references to payment processing
• find_symbol

  Exact or fuzzy search for symbols (functions, classes, etc.) with Levenshtein support

  Try:Find the User class definitionTry:Locate the process_data function
• analyze_relationships

  Analyze relationships between symbols including callers, callees, hierarchy, and imports

  Try:Show what calls the calculate_totals functionTry:Find all imports in the auth module
• get_symbol_context

  Get a 360-degree view of a symbol with all its relationships and dependencies

  Try:Show me everything about the PaymentProcessor classTry:Get full context for the handle_error function
• get_impact

  Analyze the impact of changing a symbol with risk rating

  Try:What will break if I change the database connectionTry:Show the blast radius of modifying user authentication
• detect_changes

  Map code changes to affected symbols using diff-based analysis

  Try:What symbols are affected by recent changesTry:Show which files need updating after the API refactor
• find_dead_code

  Detect code that has no callers and is potentially unused

  Try:Find functions that are never calledTry:Identify unused classes in the codebase
• find_most_complex_functions

  Identify functions with the highest cyclomatic complexity

  Try:Find the most complex functions in the projectTry:Show complexity hotspots in the payment module
• get_evolution

  Analyze how symbols have evolved over time using 6 different scoring modes

  Try:How has the user model evolvedTry:Show the most impactful changes this week
• get_timeline

  Get the full version history of a symbol

  Try:Show the timeline of the authentication systemTry:When was the payment processing last refactored
• find_bridge_symbols

  Find symbols with high betweenness centrality that connect different parts of the graph

  Try:Find bridge symbols in the codebaseTry:Identify critical connectors between modules
• list_communities

  Identify architectural modules using the Louvain community detection algorithm

  Try:Show the architectural modules in the codebaseTry:Find communities related to user management

Comparable tools

Installation

{
  "mcpServers": {
    "memtrace": {
      "command": "memtrace",
      "args": ["serve"]
    }
  }
}

FAQ

How is Memtrace different from traditional code search tools?

Memtrace creates a bi-temporal knowledge graph that maintains historical context and relationships between code elements over time, unlike traditional search tools that only provide current state.

What programming languages does Memtrace support?

Memtrace supports 20+ programming languages through Tree-sitter parsers, including mainstream languages like Python, JavaScript, Java, C++, and framework-specific languages with specialized scanners.

Compare memtrace-public with