IBM — 25 Open-Source AI Agent Skills

IBM is the author of 25 open-source AI agent skills and MCP servers spanning mcp-server, agent-tool, llm-plugin, with a combined 7,559+ GitHub stars. On AgentSkillsHub each is quality-scored (avg 56/100) and security-graded where a README exists (13 of 25) — 12 verified safe.

Top skills by IBM

• mcp-context-forge — An AI Gateway, registry, and proxy that sits in front of any MCP, A2A, or REST/gRPC APIs, exposing a unified endpoint with centralized disco (4.5k★ · quality 72/100 · security: safe)
• AssetOpsBench — AssetOpsBench - Industry 4.0: A unified benchmark and framework for building, orchestrating, and evaluating domain-specific AI agents for In (2.3k★ · quality 60/100 · security: safe)
• mcp — A collection of Model Context Protocol (MCP) servers, clients and developer tools by IBM. (401★ · quality 57/100 · security: safe)
• ibmi-mcp-server — MCP server for IBM i systems (80★ · quality 66/100 · security: safe)
• vakra — A Benchmark for Evaluating Multi-Hop, Multi-Source Tool-Calling in AI Agents (68★ · quality 60/100 · security: safe)
• responsible-prompting-api — Responsible Prompting is an LLM-agnostic tool that aims at dynamically supporting users in crafting prompts that embed responsible intention (50★ · quality 58/100)
• flexo — Framework for deploying configurable AI agents with real-time streaming and tool execution. (41★ · quality 52/100)
• data-intelligence-mcp-server — data-intelligence-mcp-server is a centralized Model Context Protocol (MCP) server that provides a unified gateway for exposing and managing (19★ · quality 68/100 · security: safe)

View IBM on GitHub · Explore audited agent skills