Repository Intelligence
Open-source codebase analysis engine for security scans, dependency graphs and tech debt
- Pricing
- Open source
- Platforms
- CLI, API, Web
- Version
- 1.0.0
- Listing updated
- today
Similar MCP servers
About Repository Intelligence
Repository Intelligence is an open-source MCP server for developer tools. It is available on CLI, API, Web, self-hosted, Linux, macOS and Windows.
The README describes Lyra Intel as a Python engine that parses code with AST analysis, maps dependency graphs, measures complexity and runs security checks for secrets, OWASP-style issues and known CVEs. It can be run locally, on a Kubernetes cluster or in a cloud environment, and it can optionally use OpenAI, Anthropic or local Ollama models for explanations and suggested fixes.
The project ships a CLI with scan and analyze commands that write results as JSON, a REST API, a web dashboard, and an MCP server that exposes analysis and search tools to Claude and other MCP clients. Results are stored in SQLite, PostgreSQL or BigQuery, depending on scale.
The README is aimed at security teams, developers taking over unfamiliar code, and engineering leads tracking technical debt. Much of the text is promotional and includes time-saved figures that are not backed by data, and the repository also contains unrelated ERC-8004 keyword material, so treat the claims with care and verify them against the code.
Tools this server provides (7)
Taken from the project documentation; the running server may differ.
- analyze-codebase
- Comprehensive code analysis with AST, dependencies, metrics
- search-code
- ML-powered semantic code search
- get-complexity
- Cyclomatic, cognitive, and Halstead complexity
- get-security-issues
- Security vulnerabilities, secrets, compliance
- discovery-scan-github
- Find new MCP crypto tools on GitHub
- discovery-analyze-repo
- Extract MCP tool definitions from repos
- discovery-run-pipeline
- Full discovery + analysis + submission
What it does
- Multi-language AST parsing
- Parses source code in several languages to extract structure, functions and classes.
- Dependency graph analysis
- Builds import relationships and flags circular dependencies between modules.
- Security scanning
- Checks for hardcoded secrets, injection patterns and OWASP-style issues, with custom rules supported.
- Complexity metrics
- Reports cyclomatic, cognitive and Halstead complexity for code.
- Optional AI explanations
- Can use OpenAI, Anthropic or local Ollama models to explain code and suggest changes.
- Flexible deployment modes
- Runs as a local process, in distributed mode or in a cloud configuration, with Docker and Kubernetes manifests.
Good for
- Scanning a legacy codebase for vulnerabilities before planning remediation
- Mapping dependencies across a large monorepo before a framework upgrade
- Adding security and complexity checks to CI pipelines
- Giving an AI assistant access to codebase analysis through MCP
Pricing
Open source
Free and MIT licensed; self-hosted.
Repository Intelligence FAQ
How is Lyra Intel installed?
The README says to clone the repository and run pip install -e . on Python 3.9 or newer.
Which languages does the AST analyzer support?
The README lists Python, JavaScript and TypeScript, Go, Rust, Java, C++, C#, Ruby and PHP.
What license does it use?
The README states the project is released under the MIT License.
Can it run without sending code to a third party?
The README says it can run locally or in your own cloud, and that local models such as Ollama can be used for AI features.
Is Repository Intelligence free?
Yes. Repository Intelligence is open source, so you can use and self-host it for free.
What are the best alternatives to Repository Intelligence?
The closest alternatives to Repository Intelligence are AI Tool Discovery Agent, CodeInspectus, Deeprepo, CybRelay and Code-Graph-RAG.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.