The control plane for MCP in real environments Blog
Articles about The control plane for MCP in real environments Blog
Ask, Plan, Act: Agent Roles That Decide What Your MCP Agent May Do
OrcA's Ask, Plan and Act roles control what an AI agent may do with your MCP servers: explore safely, review a plan, then act. Learn why roles matter even with a capability graph, and when to use each one.
OrcA 0.4: An MCP Agent Harness in VS Code, for Any MCP Server
OrcA 0.4 is a specialised harness for MCP and API work: connect any Streamable HTTP MCP server, load tools progressively to stop context bloat, edit the harness, and see capability context and plans. Free.
How to Design OpenAPI Contracts That Guide AI Agents
Enrich your OpenAPI contract so HAPI, OrcA, and Clooney find the right tool at the right moment without bloating the context, using intent, effects, planning facts, client facts, and failure outcomes.
How to Make Intent-Based Contracts and APIs AI-Ready
Deterministic workflows are rigid and free tool calling guesses. Learn how intent-based HAPI contracts and Arazzo workflows work together so AI agents ask the right questions, respect confirmations, and re-plan after failures.
How to Build and Manage MCP-Based Agents from Your API Contracts
Build AI agents from the APIs you already have. Turn OpenAPI and Arazzo contracts into MCP tools, give the agent a transparent harness in VS Code, and manage it with audits, traffic analysis, and roles with agent platforms.
OrcA 0.3: See Every MCP Tool Call and LLM Round in VS Code
OrcA 0.3 turns VS Code into an MCP traffic analyzer: chat with your MCP servers and see every tool call, LLM round, latency and token, with no telemetry.
OrcA for VS Code: From OpenAPI and Arazzo to MCP Server, Without Leaving Your Editor
OrcA is the new VS Code extension of the HAPI MCP Stack: discover OpenAPI (Swagger) and Arazzo contracts, preview their MCP tools, run them locally on a free port, deploy them, and connect them to your AI assistant.
From API to MCP with Deterministic Workflows
AI agents can interpret a user's intent, but production workflows need deterministic execution. Learn how OpenAPI, Arazzo, HAPI Workflows, and MCP connect intent to governed API outcomes without shadow logic.
HAPI Capability Graph and Jev: Safer MCP Planning for AI Agents
HAPI Capability Graph and Capability Planning help AI agents discover the right MCP tools, identify missing information, and propose the next safe step. Add optional Jev from TypeSafe AI for structured semantic ranking.