Semantic code intelligence MCP server that cuts AI token usage by ~94% with pre-computed knowledge graphs.
qartez-mcp is easy to set up with strong trust signals. Check agent compatibility and use-case fit before adding it to your workflow.
Repository setup guidance
gh repo view kuberstar/qartez-mcp --webOpen the official README and confirm the supported install method.
Add the server entry to your MCP client config.
Restart your agent and verify that the server tools appear.
Qartez is a tool that helps AI coding assistants understand your codebase without reading every file. It pre-builds a map of your code's symbols, imports, and dependencies, so the AI can answer questions and make changes using far fewer tokens. This saves money and speeds up development.
Qartez MCP is a Model Context Protocol server that provides semantic code intelligence for AI coding agents like Claude Code. Instead of relying on traditional tools like grep and find, Qartez pre-computes a knowledge graph of your repository containing symbols, imports, call edges, blast radii, PageRank, git co-change, and cyclomatic complexity. This allows AI agents to query the codebase directly, reducing token usage by approximately 94% compared to reading files line by line. The server exposes 37 MCP tools covering project mapping, symbol search, impact analysis, modification guard, and more. It supports 37 programming languages and is designed from the ground up for consumption by language models, not humans. Qartez includes three binaries: qartez (the MCP server), qartez-guard (a modification guard that prevents breaking changes), and qartez-setup (for easy installation and IDE configuration). It runs on macOS, Linux, and Windows, with pre-built binaries for x86_64 and arm64 architectures.
Strong trust signals; still review the README and permissions before production use.
Last commit was about 61 days ago.
56 GitHub stars indicate community interest.
0 open issues signal maintenance load.
NOASSERTION license detected.
AI agents can quickly find symbol definitions and references across a large codebase without scanning files.
Developers can perform impact analysis before making changes, identifying all files that might break.
Automated refactoring with confidence, knowing the full dependency graph and blast radius of changes.
Reducing token costs for AI coding assistants by replacing file reads with structured queries.
Onboarding new team members by providing a semantic map of the codebase for AI-assisted exploration.
The tool requires access to the entire repository to build its index, which may include sensitive code.
Pre-computed indexes are stored locally; ensure proper access controls if used in shared environments.
Modification guard may block changes that are actually safe if the analysis is incomplete.
56
Stars
7
Forks
0
Issues
NOASSERTION
License
Local codebase intelligence CLI and MCP server for AI coding agents with change-safety gates and audit evidence.
An offline MCP server that indexes your codebase for semantic search, code search, and git history retrieval.
Official MCP reference servers from Anthropic. Includes servers for filesystem, GitHub, Postgres, Slack, and more.
Semantic code intelligence MCP server that cuts AI token usage by ~94% with pre-computed knowledge graphs.
AI-assisted developers using Claude Code, Cursor, or other MCP-compatible agents., Engineering teams looking to reduce AI token costs and improve code understanding., Developers working on large, complex codebases with many interdependencies.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
The tool requires access to the entire repository to build its index, which may include sensitive code. Pre-computed indexes are stored locally; ensure proper access controls if used in shared environments.
Similar or complementary options to evaluate include roam-code, code-memory, MCP Server Collection.
Jun 6, 2026
3 security/trust notes recorded.
Setup difficulty is 2/5.