local-pinboard-mcp
About
A local-only, secure MCP server to connect your AI assistant to your Pinboard bookmarks
Details
- Author
- mattriffle
- Categories
- Cloud Service, Infrastructure
Jump to
Setup
Install local-pinboard-mcp in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/mattriffle/local-pinboard-mcp
Follow the installation instructions in the repository README, then restart your MCP client.
This integration uses thePinboard API(v1) with strictly typed TypeScript/Zod schemas to allow your AI assistant to manage your bookmarks, tags, and notes — without exposing credentials externally or relying on third-party services.
You can use the packaged MCPB file to get up and running quickly, or clone the repository to build your own. The source code is open, local, and auditable.
Once connected, your AI assistant can use the following tools:
List your most recent Pinboard bookmarks, optionally filtered by up to three tags.
Retrieve bookmarks matching a tag, a specific date, or a specific URL.
Retrieve all bookmarks in your Pinboard account. Rate limited to once every 5 minutes.
Add or replace a bookmark. URL and title are required. By default, replaces any existing bookmark for the same URL.
Permanently delete a bookmark by its URL. This action cannot be undone.
Return popular (site-wide) and recommended (your history) tags for a given URL.
Get all of your Pinboard tags with their usage counts. No parameters.
Rename a tag across all bookmarks, or fold it into an existing tag.
Delete a single tag from your account. Removes the tag from all bookmarks but does not delete them.
List all of your notes (without body text). No parameters. Note creation/editing is unsupported in v1 API.
Fetch a specific note including its full body text.
Returns the time of the most recent bookmark change. Use this to check if data has changed since your last fetch. No parameters.
This server usesstdiotransport exclusively — it never exposes an external HTTP port. Communication is strictly between your local machine and the officialapi.pinboard.inendpoints.
Your API token is passed to Pinboard via theauth_tokenquery parameter as required by the v1 API. To ensure absolute privacy, the server is designed to intercept and sanitize network-level error messages to prevent your token from accidentally leaking into your local Claude Desktop or Cursor logs if a network failure occurs.
Error responses are truncated safely, and input paths like tags and hex IDs are strictly verified via Zod regex before requests are made.
Drag the.mcpbfile from your downloads folder on to your AI assistant's desktop or file browser. The assistant should recognize it as an MCP server and prompt you to load it.
Ensure you have installedNode.js(version 22+ recommended) andnpmon your local machine.
- Log into yourPinboard account.
- Go toSettings → Password.
- Your API token is displayed on that page in the formatusername:HEXTOKEN.
- Copy it securely.
Clone the repository and install the required dependencies:
Copy the example environment variables file:
Open.envin your text editor and paste in your API token:
PINBOARD_API_TOKEN=username:YOURHEXTOKEN
This server works with any MCP-compatible client. Below is an example using Claude Desktop.
-
Open Claude Desktop and choose Settings → Developer → Edit Config to openclaude_desktop_config.json.
Add this to themcpServerssection, fixing the path to the compiledindex.jsfile and adding your own API token:
{ "mcpServers": { "local-pinboard": { "command": "node", "args": ["/absolute/path/to/your/local-pinboard-mcp/dist/index.js"], "env": { "PINBOARD_API_TOKEN": "username:YOURHEXTOKEN" } } } }
- Restart Claude Desktop. The new tools will appear instantly and you can start asking things like "Show me my recent unread bookmarks" or "Search my bookmarks for React tutorials".
For other MCP clients, consult their documentation on how to register a localstdioserver pointing atdist/index.jswith the required environment variables.
This project is built using TypeScript,@modelcontextprotocol/sdk, and Zod parsing.
If you are developing new tools or making adjustments:
- Compile: Always ensure you runnpm run buildafter making modifications.
- Testing: Run the Vitest test suite:
Pull requests are welcome! Please open an issue first to discuss what you'd like to change.
Navigate your Aiven projects and interact with the PostgreSQL®, Apache Kafka®, ClickHouse® and OpenSearch® services
Yunxiao MCP Server provides AI assistants with the ability to interact with the Yunxiao platform.
Get prescriptive CDK advice, explain CDK Nag rules, check suppressions, generate Bedrock Agent schemas, and discover AWS Solutions Constructs patterns.
This AWS Labs Model Context Protocol (MCP) server for CloudTrail enables your AI agents to query AWS account activity for security investigations, compliance auditing, and operational troubleshooting.
Core AWS MCP server providing prompt understanding and server management capabilities.
Analyze CDK projects to identify AWS services used and get pricing information from AWS pricing webpages and API.
Query and analyze your Axiom logs, traces, and all other event data in natural language
Manage and interact with Microsoft Azure services.
Bastion: External Attack Surface Monitoring
Ask your AI assistant about your attack surface: run scans, catch expiring certificates and domains, triage findings, and generate reports.
Agent-ready global image CDN that AI agents can install and operate through MCP.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
