Awesome GitHub RepositoriesCode Knowledge Graphs

Codegraph is a local codebase indexer and static analysis graph database that serves as a context provider for AI agents. It parses multiple programming languages into a searchable knowledge graph of symbols and dependencies, exposing these relationships to AI tools through the Model Context Protocol. The project distinguishes itself by aggregating relevant code snippets and symbol flows to reduce token usage for large language models. It automates the configuration of server settings and steering instructions across various AI agent platforms and command line editors to enable automatic code

Headroom is an AI gateway proxy and token optimizer designed to reduce the cost and latency of large language model interactions. It functions as an intermediary that intercepts traffic between clients and providers to apply context compression, request routing, and format translation. The system differentiates itself through a Model Context Protocol server implementation that delivers compression and retrieval tools to compatible AI hosts. It employs a content-aware compression pipeline and tiered importance scoring to trim redundant data from logs and tool outputs while preserving essential

CodeQL is a semantic code analysis engine and vulnerability scanning tool that treats source code as data. It utilizes a static analysis query language to define complex patterns and security vulnerabilities within a code graph database. The system represents source code as a relational database, enabling the execution of structural queries and data flow analysis. This approach allows for the detection of security flaws and coding errors across large-scale repositories. The tool provides capabilities for automated code auditing, static analysis security testing, and custom vulnerability dete

Qodo Cover ist eine Engineering-Governance-Plattform und ein KI-gestützter Assistent für automatisierte Code-Reviews und Unit-Test-Generierung. Es nutzt einen Wissensgraphen auf Basis des abstrakten Syntaxbaums (AST), um Abhängigkeiten und architektonische Beziehungen abzubilden, wodurch Pull Requests analysiert und organisatorische Coding-Standards durchgesetzt werden können. Das System zeichnet sich durch eine Multi-Agenten-Analyse-Pipeline aus, die architektonische Schlussfolgerungen zieht und Fehler identifiziert, die über das unmittelbare Diff hinausgehen. Es verfügt über einen Model-Context-Protocol-Server, um Codebase-Intelligenz für externe Tools verfügbar zu machen, und kann Durchsetzungsregeln automatisch weiterentwickeln, indem es aus historischen Pull-Request-Entscheidungen lernt. Die Plattform bietet umfassende Funktionen für das Wissensmanagement der Codebase, einschließlich Deep-Research-Ausführung, semantischer Abfragen und System-Abhängigkeits-Mapping. Sie enthält zudem Werkzeuge zur iterativen Unit-Test-Generierung zur Erhöhung der Code-Abdeckung sowie automatisierte Remediation zur direkten Anwendung von Fixes auf Pull Requests. Bereitstellungsoptionen umfassen Multi-Tenant-SaaS, Single-Tenant oder vollständig On-Premises-Installationen.

Potpie is an LLM codebase analysis platform and multi-agent orchestration framework designed to act as an AI software engineer. It parses repositories into a structured code knowledge graph, enabling AI agents to perform multi-hop reasoning, dependency tracing, and grounded technical analysis across large codebases. The system distinguishes itself through a spec-driven development framework where agents generate detailed technical specifications and architecture plans before implementing multi-file code changes. It utilizes a durable execution engine to coordinate specialized AI personas for

CodeGraphContext is a code graph indexer and visualization tool that analyzes source code to build graphs of functions, classes, and inheritance relationships. It functions as a Model Context Protocol server, providing a structured codebase index to AI assistants for context retrieval and natural language querying. The project features an interactive web interface that uses force-directed layouts to visualize code dependencies and symbols. To accelerate the setup of large projects, it supports the import of pre-calculated knowledge bundles for popular repositories. The system provides capabi