Our take
Quietno commits in the last 90 days- Activity
- 32
- Community
- 4
- Issues
- 21
- Pull requests
- 21
+0 stars in the last 10 days
- Fewer contributors than 92% of the projects we track
Ferret MCP offers a useful MCP-based route from repository structure to AI-assisted explanations, but its quiet status and dependence on one contributor make it a project to evaluate before building a critical workflow around it.
Good fit if
- You want repository analysis available inside an MCP client and can run Python 3.12 or newer.
- Static analysis is enough for your needs, or you can configure an LLM for deeper reports and questions.
- You are exploring or prototyping, and can check the output against the code when accuracy matters.
Look elsewhere if
- You need a project with recent ongoing maintenance or a broader contributor base. The repository is marked quiet, has had no commits in the last 90 days, and its work is concentrated in one contributor.
- You need a reported test suite or more evidence of ongoing issue and pull-request handling before adoption.
- Your workflow cannot supply an LLM provider for the AI-powered tools and static results do not meet your needs.
All health signals
| Last commit | 2026-04-03 (6 months ago) |
|---|---|
| Commits, last 90 days | 0 |
| Releases, last 12 months | 1 (latest v0.1.1, 2026-04-03) |
| Contributors | 1 |
| Issues closed, last 90 days | 0 |
| Pull requests merged, last 90 days | 0 |
| Project age | 6 months |
Checked on 2026-10-09 with the GitHub API.
Overview
Ferret MCP is a Python server that exposes codebase analysis tools to MCP-compatible assistants. It combines static inspection of a repository with optional LLM interpretation, helping developers understand an unfamiliar project without manually tracing every module and interface.
Static tools work without an LLM; deeper reports and repository questions require configuring an LLM provider. The project is aimed at developers and coding assistants that need repository context, rather than teams looking for a hosted analysis service.
Key Features
- Summarizes repository languages, structure, entry points, and configuration files.
- Maps external dependencies and internal imports, including core modules.
- Analyzes architectural layers, patterns, conventions, testing, and error handling.
- Extracts interface details such as REST endpoints, MCP tools, CLI commands, and exports.
- Combines static results into a comprehensive extraction report.
- Offers LLM-powered deep reports and questions grounded in codebase context.
- Supports Anthropic and OpenAI-compatible providers, including local endpoints such as Ollama.
Use Cases
- A developer joining an unfamiliar Python or other-language repository can use the static tools to locate modules, dependencies, and entry points before reading code in depth.
- A maintainer preparing a project handoff can generate an architecture and API overview to orient the next contributor.
- A coding assistant user can ask repository-specific questions when static summaries alone do not explain data flow or design choices.
- A reviewer assessing a codebase can use the report as an initial map of patterns and potential risks, then verify findings against the source.
What you need
Detected in the repository
- Python >=3.12 (from pyproject.toml)
- Automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Install and run with uvx ferret-mcp. Add it to an MCP client, and configure an LLM API key if you want to use the deep and ask tools. See the README for client-specific setup and configuration.
Alternatives
- codebase-memory-mcp: It builds a persistent graph of code structure for focused navigation queries, rather than generating static analysis and optional LLM reports.
- GitNexus: It maps code relationships into a knowledge graph for tracing execution flows and change impact, rather than producing broad codebase analysis reports.
- antivibe: It is a Claude Code skill for learning guides and architectural audits, rather than an MCP server for static analysis and optional LLM reports.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| ferret-mcp | Python | MIT | 0 | Quiet |
| codebase-memory-mcp | C | MIT | 45.7k | Active |
| GitNexus | TypeScript | Other | 47.7k | Active |
| antivibe | Shell | MIT | 1.1k | New |
Considerations
- The project is marked quiet, with no commits in the last 90 days. Its single contributor accounts for all reported contributions, so maintenance continuity is a consideration.
- Repository metadata reports CI but no test suite. The measured activity also shows no recent issue closures or merged pull requests, so there is limited recent activity to assess.
- Python 3.12 or newer is required. The static tools need no LLM, but
deepandaskrequire a configured provider; analysis quality depends on the selected model. - The README notes that only the root
.gitignoreis parsed, scans are capped at 15,000 files, and file-content analysis skips files over 512 KB. - The project is MIT-licensed. Review the license terms when incorporating or redistributing it.
Found this useful?
Share it with someone who would like ferret-mcp.