Tool Store

Persistent memory for AI coding tools. Give any MCP-compatible AI tool memory that persists across sessions — decisions, conventions, bugs, architecture, and context stored as human-readable Markdown with semantic search.

Web and local search using Brave's Search API. Supports both web search and local business search with flexible result filtering.

Connect to PostgreSQL databases. Read-only access with schema inspection and query execution. Supports connection strings.

Interact with Slack workspaces. Read messages, post to channels, manage reactions, and search conversations via the Slack API.

Secure file operations with configurable access controls. Read, write, search, and manage files within allowed directories.

Interact with GitHub repositories. Create issues, manage pull requests, search code, manage branches, and more via the GitHub API.

Browser automation using Puppeteer. Navigate pages, take screenshots, click elements, fill forms, and execute JavaScript in the browser.

Neural search engine for AI. Search the web with meaning-based queries and get clean, LLM-ready content from any webpage.

Search and read files from Google Drive. Supports Google Docs, Sheets, and Slides with automatic format conversion.

Interact with GitLab repositories and projects. Manage issues, merge requests, pipelines, and repository content through the GitLab API.

Search for places, get directions, and geocode addresses using the Google Maps Platform API. Location-aware AI interactions.

Retrieve information from AWS Knowledge Bases using Amazon Bedrock Agent Runtime. Semantic search across your AWS knowledge base.

A tool for dynamic, reflective problem-solving through a structured thinking process. Break down complex problems step-by-step with revision support.

MCP test server that exercises all MCP features — tools, resources, prompts, and sampling. Useful for testing MCP client implementations.

Interact with SQLite databases. Execute queries, analyze business data, and create schema visualizations from local database files.

Fetch web pages and convert HTML to Markdown for LLM consumption. Supports robots.txt, custom user-agent, and content extraction.

A basic MCP server providing knowledge graph-based persistent memory using a local JSON storage. Part of the official MCP servers collection.

Interact with Sentry for error tracking and performance monitoring. Search issues, get stack traces, and analyze error patterns.