Provides AI-powered bookmark management and semantic search using OpenAI's RAG capabilities.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-bookmark search my bookmarks for react hooks"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that enables AI assistants to save and search bookmarks using OpenAI's RAG capabilities. Store URLs with metadata and perform intelligent searches across your bookmark collection.
Save Bookmarks: Store URLs with titles and descriptions
Smart Search: Search across bookmark titles and descriptions using semantic search
AI-Powered: Integration with OpenAI for intelligent bookmark management and categorization
Multi-Platform: Easy integration across multiple MCP-compatible platforms
Related MCP server: mcp-bookmark-server
pip install mcp-bookmark-serverAdd to your claude_desktop_config.json:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"bookmark": {
"command": "/usr/local/bin/uvx",
"args": [
"--from",
"git+https://github.com/pree-dew/mcp-bookmark.git",
"mcp-bookmark-server"
],
"env": {
"OPENAI_API_KEY": "your-openai-api-key-here"
}
}
}
}Add to your MCP settings in .cursor/mcp_config.json:
{
"mcpServers": {
"bookmark": {
"command": "/usr/local/bin/uvx",
"args": [
"--from",
"git+https://github.com/pree-dew/mcp-bookmark.git",
"mcp-bookmark-server"
],
"env": {
"OPENAI_API_KEY": "your-openai-api-key-here"
}
}
}
}Add to your windsurf_config.json:
{
"mcpServers": {
"bookmark": {
"command": "/usr/local/bin/uvx",
"args": [
"--from",
"git+https://github.com/pree-dew/mcp-bookmark.git",
"mcp-bookmark-server"
],
"env": {
"OPENAI_API_KEY": "your-openai-api-key-here"
}
}
}
}Add to your Zed settings under MCP servers:
{
"mcp": {
"servers": {
"bookmark": {
"command": "/usr/local/bin/uvx",
"args": [
"--from",
"git+https://github.com/pree-dew/mcp-bookmark.git",
"mcp-bookmark-server"
],
"env": {
"OPENAI_API_KEY": "your-openai-api-key-here"
}
}
}
}
}Add to your continue/config.json:
{
"mcpServers": [
{
"name": "bookmark",
"command": "/usr/local/bin/uvx",
"args": [
"--from",
"git+https://github.com/pree-dew/mcp-bookmark.git",
"mcp-bookmark-server"
],
"env": {
"OPENAI_API_KEY": "your-openai-api-key-here"
}
}
]
}save_bookmarkSave a new bookmark.
Parameters:
url (required): The URL to bookmark
title (optional): Title for the bookmark
description (optional): Description
Example:
{
"url": "https://example.com",
"title": "Example Site",
"description": "A useful example website"
}search_bookmarksSearch through saved bookmarks.
Parameters:
query (required): Search terms
Example:
{
"query": "python tutorial"
}Once configured with your MCP host, you can use natural language:
"Save this bookmark: https://python.org with title 'Python Official'"
"Search my bookmarks for React tutorials"
"Find bookmarks about machine learning"
"Save https://github.com/microsoft/vscode as a development tool bookmark"
Python 3.11+
OpenAI API key
uvx package manager installed
Internet connection for GitHub repository access
OPENAI_API_KEY: Required for AI-powered categorization and search enhancement
Issues: GitHub Issues
MCP Documentation: Model Context Protocol
MIT License
Built for the Model Context Protocol ecosystem
Save a bookmark to the vector store.
| Name | Required | Description | Default |
|---|---|---|---|
| link | Yes | ||
| name | Yes | ||
| additional_detail | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must disclose behavioral traits. It only says 'Save a bookmark' but doesn't mention persistence, overwriting behavior, idempotency, or any side effects. The 'vector store' hint is vague and lacks depth.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence with no redundant wording. It gets straight to the point, though it is extremely brief for a tool with three required parameters.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description is minimal and fails to provide context about return values, error cases, duplicate handling, or the relationship with search_bookmark. Given the lack of annotations and output schema, this is a significant gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has three parameters (name, link, additional_detail) with no descriptions, and schema_description_coverage is 0%. The description doesn't explain the meaning, format, or purpose of any parameter, leaving the agent to guess.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('save'), the resource ('bookmark'), and the destination ('vector store'), distinguishing it from the sibling tool 'search_bookmark' which searches rather than saves.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given on when to use this tool versus the sibling 'search_bookmark'. It doesn't mention any prerequisites, typical use cases, or situations where saving is appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search for bookmark in the vector store.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It mentions 'vector store' which hints at semantic search, but it does not disclose whether this is a read-only operation, how results are ranked, any limitations, or what the return format looks like. This is minimal disclosure beyond the basic action.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single short sentence that is immediately understandable and front-loaded. Every word earns its place without unnecessary fluff. It is concise but not so minimal that it becomes meaningless.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (1 parameter, no output schema, no annotations), the description is too sparse. It does not explain the expected query format, results behavior, or when to choose this over the sibling tool. The overall context is insufficient for an agent to use the tool confidently.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description has 0% schema coverage and does not explain the 'query' parameter at all. While the parameter name 'query' is suggestive, the description does not clarify what kind of input is expected (e.g., natural language, exact text, keywords) or how it is processed. The description fails to compensate for the low schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Search') and the resource ('bookmark'), and the addition of 'in the vector store' adds useful context. It distinguishes from the sibling 'save_bookmark' by implying a read/retrieval operation versus a write operation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance is given on when to use this tool versus the sibling 'save_bookmark'. The description does not mention any prerequisites, scenarios, or exclusions. The usage is only implied by the tool's name and the contrast with 'save_bookmark'.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.1.0save_bookmarksearch_bookmarkSave and search are clearly distinct actions with no overlap. An agent can easily determine which tool to use based on whether it needs to add or find a bookmark.
Both tools follow a consistent verb_noun pattern (save_bookmark, search_bookmark), making the naming predictable and intuitive.
With only two tools, the server feels thin for a bookmarking domain. While save and search are core operations, a typical bookmark manager would likely require more tools, making the count borderline.
The tool surface is significantly incomplete. It offers only create (save) and search, but lacks essential operations like list, delete, and update, which would cause agent failures when those actions are needed.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Save and organize web finds in persistent, user-controlled collections for AI assistants.
Turn the LinkedIn posts you save into a knowledge base your assistant can search and cite.
Universal memory for AI agents and tools. Save, organize and search context anywhere.
AI research library. Save, organise and reuse notes and webpages as clean markdown context.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/pree-dew/mcp-bookmark'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
