$ loading_
Enable Claude Code agents to ask humans, request approval, and report status.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "cware-hil-mcp" yet — see the docs or source repo.
Before deleting tables in the production database, use cware-hil-mcp to send a human approval request with the reason, impact scope, and rollback plan, then wait for explicit approval before continuing.
The tool sends a structured approval request to a human and returns a continue status after approval is granted.
If the API field naming convention in a coding task is unclear, use cware-hil-mcp to ask a human 3 specific questions, then turn the answers into actionable implementation requirements.
The tool sends questions to a human, gathers responses, and produces clear constraints for the next development steps.
During a project refactor, use cware-hil-mcp to report progress to a human after each phase, including completed modules, discovered risks, and the next plan.
The tool continuously updates a human on task status, helping them track progress and step in for decisions.
Query Claude Code transcript analytics for cost, safety, audit, and efficiency insights.
Coordinate multiple AI coding agents to build, review, and remember together.
Connect and operate multiple MCP-enabled CMS platforms through one natural-language endpoint.
Unifies MCP clients with tool routing, memory, and automation flows.
Deploy a remote MCP server on Cloudflare Workers for AI client connections.
Connect Claude Code to OpenClaw agents to query, invoke, and monitor them.