Obsidian Regulatory MCP

io.github.richbilePublished by GitHub account richbilev0.1.0Updated Nov 2, 2026

Verified, tier-0 regulatory data for AI across 850+ official sources and 50+ jurisdictions.

VerifiedStreamable HTTPWeb executableWeb Search & ScrapingKnowledge & MemoryBusiness & Commerce

Overview

AI-generated overview

Remote MCP server giving assistants verified regulatory search across 850+ official sources in 50+ jurisdictions, with source, date and status on each result.

What you can do
  • Search verified official regulations semantically, filtered by jurisdiction, framework and status, for obligations, deadlines and legal status
  • Look up enforcement and recall events such as product recalls, safety alerts, bans and official enforcement actions
  • Check standards and guidance lifecycles, including current, superseded, draft or withdrawn editions, for life sciences and medtech
  • Run keyword searches across press and media coverage of regulatory topics for background color
When to use it
Useful when an assistant answers compliance, legal or regulatory questions and needs citable official sources rather than model recall, for example tracking what changed in CSRD reporting, checking REACH restrictions, or finding MDR transition deadlines.
Requirements
A remote MCP endpoint added as a custom connector; no API key or headers. Sign-in with an Obsidian account via OAuth 2.1 in the browser on first use. A free tier is available without a credit card; results are limited to the jurisdictions and industries authorized on the account.
Before you install
Requires an Obsidian account sign-in through OAuth, so the provider sees your identity and which jurisdictions and industries your account is scoped to. The free tier is capped at 100 verified requests per month, and paid tiers are not described here. Regulatory answers should still be checked against the cited official source before being relied on.

Installation

In SourceWeft

  1. Open Obsidian Regulatory MCP in the dashboard and add it to a workspace.
  2. Connect the required credentials. They are stored privately and never shown on public pages.
  3. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.richbile%2Fobsidian-regulatory-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.

{
  "mcpServers": {
    "obsidian-regulatory": {
      "type": "http",
      "url": "https://mcp.obsidianri.com/mcp"
    }
  }
}

README

Obsidian Regulatory MCP

The verified, tier-0 regulatory data layer for your AI. Connect Claude, ChatGPT, or Cursor to official regulatory data, with a source, date, and status on every answer.

[Website] [MCP] [Free tier] [License]

Obsidian Regulatory MCP is a Model Context Protocol server that gives AI assistants real-time access to verified regulatory data from 850+ official, tier-0 sources across 50+ jurisdictions (Chemicals, ESG, Life Sciences). It is the missing data layer that stops AI from hallucinating on regulation.

  • 🌐 Website and access: https://obsidianri.com/mcp
  • 🟢 Status: live. Sign in with Obsidian (OAuth 2.1), no API key. Free tier, no credit card.

Why this exists

Frontier models reason brilliantly but guess on regulation: hallucinated citations, stale texts, and no way to tell a draft from a law in force. In compliance, a confident wrong answer is a liability.

The fix is not a smarter model, it is a verified source of truth the model can call. You already have the intelligence. Obsidian brings the data.

Read the thesis: Agentic Regulatory Intelligence · Why AI Hallucinates on Regulatory Questions · RAG vs MCP · What Tier-0 Regulatory Data Means

Works with

Claude (Desktop and Claude Code), Cursor, and any MCP-compatible client. ChatGPT and Perplexity as MCP support rolls out across them.

Quickstart

No API key to copy. Add the connector URL; on first use your assistant opens a browser to sign in with Obsidian (OAuth 2.1). New accounts get the free tier automatically. Full walkthrough: How to Connect a Regulatory MCP to Claude.

Claude Desktop / Cursor

Add a custom connector pointing at:

https://mcp.obsidianri.com/mcp

No auth headers needed: the client discovers the OAuth flow automatically and prompts you to sign in.

Claude Code

bash
claude mcp add --transport http obsidian-regulatory https://mcp.obsidianri.com/mcp

Then authorize in the browser when prompted. See also examples/ for raw config.

Auth is OAuth 2.1 (sign in with your Obsidian account). The free tier needs no credit card.

What you get

  • 850+ official, tier-0 sources across 50+ jurisdictions (EU, US, UK, CH, APAC, and more)
  • Every result returned with its source, date, jurisdiction, framework, and status
  • Coverage of REACH, CLP, TSCA, CSRD, ESRS, EU Taxonomy, CS3D, MDR, IVDR, FDA 21 CFR, the AI Act, and more
  • Free tier: 100 verified requests per month, no credit card

Example tools

The server exposes structured tools your assistant calls in natural language:

ToolWhat it does
regulatory_searchPrimary tool. Semantic search across verified, tier-0 official regulations, filtered by jurisdiction, framework, and status. Use it for obligations, deadlines, legal status (in force / proposed / repealed) and for what changed recently. Every result carries its official source, date, and status.
search_enforcementEnforcement and recall events: product recalls, safety-gate alerts, sales or import bans, market withdrawals, official enforcement actions (for example FDA warning letters). What actually happened on the ground, with the official source and date.
standards_lookupStandards and guidance with their version lifecycle: current applicable edition, what it superseded, status (current / superseded / draft / withdrawn). Families such as ISO, IEC, MDCG, Team-NB, IMDRF, ICH, WHO, EU MDR Common Specifications. Available for industries that maintain a standards register (today: life sciences / medtech).
search_newsSecondary tool. Keyword search across press and media coverage of regulatory topics, returning title, source, and date. Use it for reporting color after answering from regulatory_search.

Results are scoped to the jurisdictions and industries authorized on your Obsidian account.

Example prompts

"What changed on CSRD reporting this month? EU only, top 3, with source and date.""Is this substance on the REACH Candidate List, and restricted in the EU and US?""What MDR transition deadlines apply to a Class IIb device?"

More: Regulatory Research Prompts That Work · Building an Agentic Compliance Workflow

Get access

Just add the connector and sign in with your Obsidian account, no API key to copy. Free tier, no credit card. Learn more at obsidianri.com/mcp.

About

Built by Obsidian Regulatory Intelligence, which continuously monitors official regulatory sources worldwide for compliance, legal, and strategy teams.

License

MIT (documentation and configuration examples). See LICENSE.

Source: README.md at commit 4baeb2d

Tools

0
This server asks you to sign in, so SourceWeft could not read its tools without an account and has not indexed them. What a connected account sees can depend on that account and its permissions.

Version history

1
  1. v0.1.0LatestSep 16, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts