AI Distiller (aid)

Distill large codebases into AI-friendly context — public APIs, types, and structure, 90-98% smaller.

Unverified

stdio (local)

No auth

View repo 160 npm

Add to your client

Copy the config for your MCP client and paste it into its config file.

Install / run

claude mcp add aid -- npx -y @janreges/ai-distiller-mcp

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ai-distiller-aid": {
      "command": "npx",
      "args": [
        "-y",
        "@janreges/ai-distiller-mcp"
      ]
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

Node.js with npx (the MCP server is installed via `npx -y @janreges/ai-distiller-mcp`)
An MCP-compatible client (Claude Code/Desktop, Cursor, Windsurf, VS Code)

About AI Distiller (aid)

AI Distiller solves the AI context problem: AI models have limited context windows and cannot comprehend an entire large codebase, so they grep for keywords and guess interfaces, producing code full of errors. AI Distiller extracts only the essential information (public API signatures, types, structure) AI needs to write code correctly on the first try, typically generating context that is only 5-20% of the original source volume. The MCP server exposes this capability to AI agents directly, plus specialized tools that generate pre-configured analysis prompts (security, refactoring, performance, bug-hunting, diagrams, docs) packaged with distilled code for the agent to execute.

Tools & capabilities (15)

distill_file

Extract structure (public APIs, types, signatures) from a single file.

distill_directory

Extract structure from an entire directory of source code.

list_files

Browse directories with file statistics.

get_capabilities

Get information about AI Distiller capabilities.

aid_hunt_bugs

Generate bug-hunting prompts with distilled code.

aid_suggest_refactoring

Create refactoring analysis prompts.

aid_generate_diagram

Produce diagram generation prompts (Mermaid).

aid_analyze_security

Generate security audit prompts (OWASP Top 10).

aid_generate_docs

Create documentation generation prompts.

aid_deep_file_analysis

Systematic file-by-file analysis workflow.

aid_multi_file_docs

Multi-file documentation workflow.

aid_complex_analysis

Enterprise-grade analysis prompts.

aid_performance_analysis

Performance optimization prompts.

aid_best_practices

Code quality and best practices prompts.

aid_analyze

Core analysis engine — direct access to all AI actions for custom workflows.

What this server can do

AI Distiller (aid) provides tools for these capabilities — tap one to see every MCP server that does the same:

Search documentation

When to use it

Give an AI agent complete knowledge of a large codebase's public interfaces, parameter types, and return types so it writes correct code on the first try instead of guessing.
Compress massive frameworks (e.g. Django ~10M tokens to ~256K) so the entire project fits in a single AI conversation, reducing hallucination and API cost.
Generate ready-to-execute security audit, refactoring, bug-hunting, performance, or documentation prompts bundled with distilled code.
Zoom AI agents into specific modules (auth, API, database) with focused distilled context instead of overwhelming them with irrelevant code.

Security notes

AI Distiller extracts code structure that may include function/variable names revealing business logic, API endpoints and internal routes, type information, comments/docstrings (unless stripped with --comments=0), and file paths revealing project structure. Always review output before sending to external services; consider running a secrets scanner on your codebase first. The tool runs 100% locally and makes no network connections of its own.

AI Distiller (aid) FAQ

Is my code sent anywhere?

No. AI Distiller runs 100% locally. It only extracts and formats your code structure — you decide what to do with the output. The tool itself makes no network connections.

Which programming languages are supported?

Currently 12+ languages via tree-sitter: Python, TypeScript, JavaScript, Go, Java, C#, Rust, Ruby, Swift, Kotlin, PHP, and C++. All parsers are bundled in the binary — no external dependencies.

Does AI Distiller perform the analysis itself?

No. It generates specialized analysis prompts combined with distilled code; AI agents (Claude, Gemini, ChatGPT) then execute those prompts. For large codebases you can copy the output into tools with very large context windows like Gemini.

Can it handle very large repositories?

Yes — it has been tested on repositories with 50,000+ files. Parallel processing scales near-linearly with CPU cores, and large files are processed in streaming chunks to keep memory bounded.

#code-context #code-distillation #tree-sitter #ast #code-analysis #token-reduction #cli #go

Alternatives to AI Distiller (aid)

Compare all alternatives →

Playwright MCP Server

Developer Tools

24k

Microsoft's official browser-automation MCP using Playwright's accessibility tree (no vision model).

Featured

Verified

stdio (local)

No auth

TypeScript

12 tools

Updated 13 days agoRepo

Context7 MCP Server

Developer Tools

28k

Up-to-date, version-specific library documentation injected into your coding agent.

Verified

stdio (local)

API key

TypeScript

2 tools

Updated 17 days agoRepo

Serena

Developer Tools

12k

LSP-powered coding agent toolkit: semantic symbol search, references and structural edits.

Verified

stdio (local)

No auth

Python

11 tools

Updated 15 days agoRepo

Compare AI Distiller (aid) with:

vs Playwright MCP Server vs Context7 MCP Server vs Serena vs Mobile MCP