Packages

An MCP (Model Context Protocol) server that exposes a single tool to run GNU Make targets from a specified Makefile (and optional working directory), capturing output and enabling LLM-assisted understanding/execution of build tasks.

MCPShim provides a local daemon (mcpshimd) and CLI (mcpshim) that turn remote MCP server tools into standard command-line workflows. It manages MCP session/discovery, optionally OAuth login, and exposes tool calls via a JSON-over-UNIX-socket IPC protocol and mapped CLI flags, returning structured JSON output and maintaining local call history in SQLite.

memorious-mcp is a local Model Context Protocol (MCP) server that provides semantic long-term memory operations—store, recall (vector similarity search), and forget—using ChromaDB for persistent on-disk storage.

obsidian-web-mcp is a Python MCP server that exposes an Obsidian vault over HTTPS so remote MCP clients (e.g., Claude web/phone) can read and write Markdown files. It provides tools for reading/writing files, updating YAML frontmatter, searching (body and frontmatter), listing directories, moving/renaming, and soft-deleting to a .trash folder. It claims OAuth 2.0 (authorization code flow with PKCE), bearer-token validation per request, Cloudflare Tunnel-based exposure, path traversal protections, and atomic writes to avoid partial file states that could break Obsidian Sync.

MCP server for the Opendock Neutron API

Remind (remind-mcp) is a generalized long-term memory layer for LLM-driven agents. It ingests “episodes” of experiences, consolidates them into generalized “concepts” (with confidence/conditions/exceptions), maintains an entity/knowledge graph, and supports semantic/graph-based retrieval. It can be used via a local CLI/skills workflow with a project database, or via an MCP server for IDE/desktop agents, which also exposes a web UI and a small REST API.

searxNcrawl provides a minimal MCP server (STDIO and HTTP) plus CLI tools to search the web via SearXNG and crawl web pages/sites, extracting readable Markdown using Crawl4AI with configurable deduplication and optional authenticated crawling via Playwright storage_state.

Provides an offline Ethereum wallet signing toolkit with MCP servers (including signing-mcp-server) for EIP-191 message signing and EIP-712 typed-data signing/verification, plus related wallet/keystore utilities in the broader toolkit. Emphasizes offline/self-contained operation and uses established Ethereum cryptography libraries.

transaction-mcp-server is an offline Ethereum wallet/transaction MCP server that exposes tools to build, encode/decode, hash, and sign Ethereum transactions (legacy/Type 0, access-list/Type 1, EIP-1559/Type 2), plus related helpers like calldata parsing and transaction validation. It runs as a standalone Python package and is designed to be used by AI assistants via the Model Context Protocol (MCP).

Provides an MCP server (stdio) that wraps a TypeScript Language Server to offer LSP-backed code intelligence tools such as go-to-definition, hover, and find-references for JavaScript/TypeScript projects.

Provides an MCP (Model Context Protocol) server plus an After Effects “bridge” panel to let clients remotely control Adobe After Effects project contents—e.g., create compositions and layers, modify properties, set keyframes/expressions, and run scripts—via MCP tool calls that ultimately execute ExtendScript/JS inside After Effects.

Provides an MCP server that bridges an AI assistant (e.g., Claude Desktop, VS Code Copilot Chat) to Autodesk Fusion 360 via an add-in, enabling tool-based generation and modification of CAD models (sketch creation, modeling operations, parameter changes, and STEP/STL export).

charles-mcp is an MCP server that connects an AI agent to Charles Proxy’s Web Interface/recording data, enabling live capture reading and structured analysis of HTTP traffic (with a summary-first approach and drill-down to per-entry details). It can also analyze recorded sessions/history and provide control/status tools for the Charles environment.

Computer Use Protocol (CUP) defines a cross-platform, LLM-friendly schema for representing desktop UI accessibility trees (JSON envelope and a compact text encoding) and the canonical action vocabulary needed for AI agents to perceive and interact with UI elements. This repo provides the core schema/spec; execution is implemented by SDKs and (optionally) MCP servers in the ecosystem.

Ask any GitHub repository a question. Get source-backed answers.

A Go example implementation of a JSON-RPC 2.0 compliant MCP (Model Context Protocol) server for note management. It provides a command-line interface and a service component (intended to run over stdio), with thread-safe in-memory note storage plus example MCP prompts/tools such as adding notes and summarizing notes. Includes an error code table and guidance for running via Claude Desktop MCP server configuration.

Provides a Model Context Protocol (MCP) server that exposes Kubernetes API functionality (read-only resource inspection plus optional mutation like deleting pods and scaling deployments) over MCP transports (stdio by default and SSE).

Rust framework for building AI agents that communicate with Model Context Protocol (MCP) servers via an MCP client (HTTP and stdio) and that can also expose custom tools via an MCP server. Includes example agent integrations for Anthropic (Claude) and OpenAI, plus a web-based inspector for debugging MCP servers/tools.

Provides an MCP server that converts Markdown to Atlassian Confluence content and uploads it (create/update page) including Mermaid diagrams (rendered to PNG via kroki.io), images (uploaded as Confluence attachments), code blocks (Confluence code macro), and preserves tables/links/formatting.

motion-dev-mcp is an MCP (Model Context Protocol) server that provides offline access to Motion.dev documentation, code examples, component/API reference, and tool-assisted animation code generation/validation for React/JavaScript/Vue. It uses a local SQLite-first documentation database with full-text search (FTS5).