# Docket Lens

> Docket Lens is an open-source MCP server for legal and government. Retrieve FDA docket comments, attachments and draft-vs-final guidance for review. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted
- **Website:** https://blackswancausallabs.com/docket-lens.html
- **Repository:** https://github.com/black-swan-causal-labs/regintel-mcp
- **MCP Registry name:** com.blackswancausallabs/regintel-mcp
- **Categories:** legal and government, health and science
- **Listing updated:** 2026-11-14

## About

Docket Lens pairs an open-source retrieval server, regintel-mcp, with an agent workflow for studying FDA guidance dockets on Regulations.gov: who commented, what they asked for, and how the final guidance compared with the draft. The server only fetches, caches, extracts and diffs; it makes no model calls, and the interpretation is left to the AI model, guided by project skills.

Tools search dockets, fetch docket metadata, list notices and guidance files, page through comments (flagging stubs and duplicates without dropping any), read a comment with text extracted from its attachments, pull a whole docket into a resumable local SQLite cache, split a guidance document into numbered sections, and diff a draft against the final text with moved-paragraph detection. A corpus export writes a JSONL file for skills and subagents.

The suggested analysis has four human checkpoints: confirming sources, reviewing submitter classification, approving the codebook, and sampling theme coding and guidance comparisons. A separate Explorer, licensed PolyForm Noncommercial, presents the reviewed result. The site notes that a request reflected in final guidance shows concordance, not that the comment caused the change. A free Regulations.gov API key is needed for a new pull; the server is MIT licensed and runs locally on Python 3.11 or later.

## Features

- **Docket search and metadata:** Searches FDA dockets by keyword and returns docket details and comment counts.
- **Comments with attachment text:** Reads comments and extracts text from their attachments.
- **Resumable corpus builds:** Caches every comment and attachment locally and reports stubs, duplicates and OCR needs.
- **Draft versus final diff:** Aligns guidance sections and flags paragraph changes and moved paragraphs.
- **Deterministic retrieval:** Makes no model calls, leaving judgement to the AI client and human checkpoints.
- **Explorer output:** Feeds a separate interactive view of reviewed findings.

## Good for

- Analyse who commented on an FDA guidance docket
- Compare draft and final guidance section by section
- Extract text from comment attachments for coding
- Build a docket corpus file for skills and subagents
- Check whether final guidance reflects requests in comments

## Install

### Claude Code (Terminal)

```sh
claude mcp add regintel-mcp -- npx -y https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "regintel-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "regintel-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "regintel-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "regintel-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "regintel-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb"
      ]
    }
  }
}
```

## Tools

- `search_dockets`: Keyword search over dockets.
- `get_docket`: Docket metadata and total comment count.
- `list_documents`: Notices, draft and final guidance and meeting material with file URLs.
- `list_comments`: All comments, flagging stubs and duplicates without dropping any.
- `get_comment`: Comment body, submitter fields, attachments and extracted attachment text.
- `build_corpus`: Pulls every comment and attachment into the cache; resumable.
- `get_guidance_text`: A guidance document split into numbered sections.
- `diff_guidance`: Section-aligned diff of draft and final guidance with moved-paragraph detection.
- `export_corpus`: Writes a corpus.jsonl file for skills and subagents.

## FAQ

### What do I need for a new analysis?

A free Regulations.gov API key, Python 3.11 or later, and a compatible AI application; viewing the existing Explorer needs no installation.

### Does the server interpret the comments?

No. It fetches, caches, extracts and diffs only; the AI model does the interpretation.

### Does it prove comments caused changes?

No. The project says that a request reflected in final guidance shows concordance, not causation.

### What are the licences?

The retrieval server is MIT; the Explorer is PolyForm Noncommercial.

### How are rate limits handled?

A personal key allows 1,000 requests per hour, and a local cache answers repeat reads.

### Is Docket Lens free?

Yes. Docket Lens is open source, so you can use and self-host it for free.

### How do I add Docket Lens to Claude Code?

Run this command in your terminal: claude mcp add regintel-mcp -- npx -y https://github.com/Black-Swan-Causal-Labs/regintel-mcp/releases/download/v0.1.0/docket-lens-0.1.0.mcpb

### What are the best alternatives to Docket Lens?

The closest alternatives to Docket Lens are Tooluniverse, Oya Browser, NY LLC Publication (FileMyPublication), Later Stages and TribuJus.

## Alternatives

- [Tooluniverse](https://ai-powered.com/mcp/tooluniverse): Tooluniverse is an open-source MCP server for health and science.
- [Oya Browser](https://ai-powered.com/mcp/oya-browser): Oya Browser is a freemium MCP server for browser automation.
- [NY LLC Publication (FileMyPublication)](https://ai-powered.com/mcp/ny-llc-publication-filemypublication): NY LLC Publication (FileMyPublication) is a free MCP server for legal and government.
- [Later Stages](https://ai-powered.com/mcp/later-stages): Later Stages is a free MCP server for health and science.
- [TribuJus](https://ai-powered.com/mcp/tribujus): TribuJus is an open-source MCP server for legal and government.
- [SupplyClub Catalog](https://ai-powered.com/mcp/supplyclub-catalog): SupplyClub Catalog is a free MCP server for legal and government.
- [SynomosAI Governance MCP](https://ai-powered.com/mcp/synomosai-governance-mcp-2): SynomosAI Governance MCP is an open-source MCP server for security.
- [MedXpert 标准检索连接器](https://ai-powered.com/mcp/medxpert-2): MedXpert 标准检索连接器 is a free MCP server for legal and government.

---
Source: https://ai-powered.com/mcp/docket-lens
