A "Codebase" as an MCP Server
We built a reference MCP server that lets your editor/agent learn a codebase directly from source (signatures, types, AST, comments). Docs are optional. The reference impl runs on our open-source project’s codebase. Why we built it Docs are important, but they add another abstraction layer between your code and your users. Keeping them at the right quality is hard (especially at a startup), and LLM-generated docs are often mediocre until you invest real polish. Exposing code to the model in a structured way keeps answers grounded and current, and it’s always available. You can even…
In plain words
An MCP server that enables editors and AI agents to learn codebases directly from source code by analyzing signatures, types, abstract syntax trees, and comments. Designed for developers who want AI tools to understand their code structure without relying on separate documentation. The server keeps information current and grounded in actual code rather than maintaining separate docs, and supports switching between different branches or releases.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
We built a reference MCP server that lets your editor/agent learn a codebase directly from source (signatures, types, AST, comments). Docs are optional. The reference impl runs on our open-source project’s codebase. Why we built it Docs are important, but they add another abstraction layer between your code and your users. Keeping them at the right quality is hard (especially at a startup), and LLM-generated docs are often mediocre until you invest real polish. Exposing code to the model in a structured way keeps answers grounded and current, and it’s always available. You can even switch across releases/branches. Why share this Building this for our own repo showed us how much leverage such a tool can give and how little glue you need to ship one. We did it for ourselves, but we think every repo can (and probably should) do it too. Bigger picture In platforms/infra, “learn a new paradigm” has been an incumbent moat. LLMs + the right MCP tools can flatten that curve so more people adopt new stacks quickly and ship faster. How to try it Hosted over HTTP [2]. With Claude Code: claude mcp add -t http fenic-docs https://mcp.fenic.ai Then ask: “What is fenic?” For local and more prompts, see [1]. If you want to take a look at the data without running the pipeline and performing the processing + inference, check [3]. Links [1] Example impl + prompt starter pack + local setup: https://github.com/typedef-ai/fenic/tree/main/examples/mcp/d... [2] Hosted endpoint: https://mcp.fenic.ai [3] Generated dataset used by the MCP server: https://huggingface.co/datasets/typedef-ai/fenic-0.4.0-codeb... It’s not perfect, but it meaningfully improves the DX. We’d love feedback!
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, September 2025
the whole month →
- AS
Commerce · Sep 2025 · anycrap.shop
- TE
I made a built-from scratch Wayland Compositor to display any GUI app* in the terminal! I think there is a lot of unexplored potential in custom Wayland compositors, a lot of really cool things you can embed existing applications into! So, I started with embedding apps into the terminal because that is the easiest input/output (output is just utf-8 and I use the great `chafa` library for that, and I just read from stdin for the input). If you have any other ideas for cool Wayland compositors, let me know. I purposedly wrote 80% the app in Typescript to appeal to the most developers and…
Dev tools · Sep 2025 · github.com
- IR
Years ago I stumbled across a basic version of this concept and it stuck with me. I knew if I was ever going to take on such a project, it would need to be flawless, but without coding experience it was just another idea that would never happen. By the end of 2024, as AI coding tools exploded everywhere, I finally had a way to make it real. I started from zero knowledge and spent months collaborating with AI agents as a learning experience. Every pixel and every function went through me. The AI translated what I asked for into code, but every decision was human. I didn't use existing OS…
AI · Sep 2025 · mitchivin.com
