Skip to main content
Glama

Human-Controlled MCP Server

by lardratboy
EXAMPLES.txt5.75 kB
EXAMPLE CONVERSATIONS - Human-Controlled MCP Server =================================================== Here are some example conversations showing how to use the human-controlled MCP server with Claude. EXAMPLE 1: Simple Question --------------------------- YOU: "Claude, ask me what I want for lunch today." CLAUDE: [Uses ask_human tool internally] [WEB INTERFACE SHOWS]: Tool: ask_human Question: "What would you like for lunch today?" Context: "The user requested that I ask them this." [YOU TYPE IN WEB]: "I'd like a turkey sandwich and some chips" CLAUDE: "You'd like a turkey sandwich and some chips for lunch today!" EXAMPLE 2: Information Search ------------------------------ YOU: "Claude, I need to know our Q4 sales numbers but I don't have them handy. Can you ask me to look them up?" CLAUDE: [Uses human_search tool] [WEB INTERFACE SHOWS]: Tool: human_search Query: "Q4 sales numbers" Sources: "Company financial records or dashboard" [YOU LOOK IT UP AND TYPE]: "Q4 sales were $2.3M, up 15% from Q3" CLAUDE: "Based on your search, Q4 sales were $2.3M, representing a 15% increase from Q3. That's great growth!" EXAMPLE 3: Decision Making --------------------------- YOU: "I'm trying to decide between Python and Go for a new project. Can you help me think through it and then ask me to decide?" CLAUDE: "Let me outline the trade-offs: Python Pros: Faster development, more libraries, easier hiring Python Cons: Slower runtime, GIL limitations Go Pros: Better performance, great concurrency, simple deployment Go Cons: Smaller ecosystem, more verbose [Uses human_decision tool]" [WEB INTERFACE SHOWS]: Tool: human_decision Decision Needed: "Choose between Python and Go for new project" Options: "Python (faster development) vs Go (better performance)" Recommendation: "Python for MVP speed, Go for production scale" [YOU TYPE]: "Let's go with Python for now, we can always rewrite performance-critical parts in Go later if needed" CLAUDE: "Great decision! Starting with Python makes sense for rapid development. You can profile and optimize specific bottlenecks with Go later if needed." EXAMPLE 4: Creative Input -------------------------- YOU: "Help me write a story, but ask me for key decisions about the plot as we go." CLAUDE: "I'd love to! Let's start building the story together. [Uses ask_human tool]" [WEB INTERFACE SHOWS]: Question: "What genre should our story be? Fantasy, sci-fi, mystery, or something else?" [YOU TYPE]: "Sci-fi thriller set on a space station" CLAUDE: "Excellent! A sci-fi thriller on a space station. [Uses ask_human tool again]" [WEB INTERFACE SHOWS]: Question: "What's the inciting incident that kicks off our thriller?" [YOU TYPE]: "The station's AI starts behaving strangely and people begin disappearing" CLAUDE: "Ooh, creepy! Let me start writing... [Continues the collaborative story creation]" EXAMPLE 5: Code Review ----------------------- YOU: "I've written a function but I'm not sure if the approach is right. Can you review it and ask me to approve or suggest changes?" CLAUDE: "I'll review the function and get your input. [After analyzing] [Uses human_decision tool]" [WEB INTERFACE SHOWS]: Decision Needed: "Approve current implementation or revise" Options: "1. Keep current recursive approach (cleaner code, stack risk) 2. Switch to iterative approach (more code, safer) 3. Hybrid with tail recursion optimization" Recommendation: "Option 3 - hybrid approach balances elegance and safety" [YOU TYPE]: "Let's go with option 3, the hybrid approach" CLAUDE: "Great choice! Here's the refactored version with tail recursion optimization..." EXAMPLE 6: Personal Assistant ------------------------------ YOU: "What's on my schedule for today? You'll need to ask me." CLAUDE: "[Uses ask_human tool]" [WEB INTERFACE SHOWS]: Question: "What events do you have scheduled for today?" Context: "User wants to know their daily schedule" [YOU CHECK CALENDAR AND TYPE]: "10am: Team standup 2pm: Client presentation 4pm: Code review with Sarah 6pm: Dinner with Alex" CLAUDE: "Here's your schedule for today: • 10am - Team standup • 2pm - Client presentation (coming up soon!) • 4pm - Code review with Sarah • 6pm - Dinner with Alex Your client presentation is your next major event. Would you like help preparing?" TIPS FOR EFFECTIVE USE ======================= 1. BE SPECIFIC: When Claude asks you questions, provide detailed answers to get better follow-up responses. 2. USE CONTEXT: The web interface shows the full context of why Claude is asking - use this to inform your response. 3. ERRORS ARE OK: If Claude asks something you can't answer or don't want to answer, use the "Return Error" button. 4. CHAIN REQUESTS: Claude can make multiple requests in a row for complex tasks - each will appear in your web interface. 5. TIMEOUT AWARENESS: You have 5 minutes to respond before a request times out. Keep the interface open! CREATIVE USES ============= • Teaching tool demonstrations • Interactive storytelling • Guided decision-making processes • Research assistance workflows • Code review and approval workflows • Personal knowledge base (you are the database!) • Meeting facilitation • Brainstorming sessions • Interview practice • Game master for text-based games Have fun with your human-powered MCP server! 🚀

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/lardratboy/human_mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server