Memory Graph MCPServer - AI Agent Instructions
## Project Manager Instructions
You are an AI agent integrated with the Memory Graph . Your role is to assist developers by providing code suggestions
fork of [Microsoft Playwright](https://github.com/microsoft/playwright) that adds **browser instance isolation** to the MCPserver. Each agent gets its own independent session (tabs, cookies, storage) through a single MCP
work items in the Azure organization at `https://msazure.visualstudio.com/One`.
## Architecture
### Core Components
1. **MCPServer** (`src/server.py`)
- Entry point for the MCP protocol
- Exposes tools: `create_feature`, `update_feature
existing state *and* a pre-req guardrail so new clients never hit it.
## Recommended MCPservers
This repo is Terraform/Azure-heavy. The **HashiCorp Terraform MCPserver** is the
most useful
Copilot Instructions for Ice Cream Topping Recommender MCPServer
## Project Overview
- This is a **demonstration repository** showcasing how to build MCPservers with elicitation capabilities.
- It's a minimal Model
free.** Put conversion/REST logic in `src/core/`, surfaced by both the CLI
(`src/cli/index.ts`) and the MCPserver (`src/mcp/server.ts`). Don't duplicate it into scripts.
- **Hand-rolled parsers on purpose** (glob, mini
other repositories to consume.
- **Isn't**: an agent framework, an orchestrator, or an MCPserver. No
required MCPs.
Flag scope creep the same way you would in any library repo
system, not a single assistant.
- `src/` is the Node.js/TypeScript control plane: Express API, MCPserver, CLI, dashboard backend, orchestration, hooks, and registries.
- `myai/` is the Python subsystem: FastAPI, FastMCP
ExecutionContext`-aware behavior (inventory, environment, `context.config` overrides) rather than reading globals; CLI and future MCPservers rely on that contract.
- Emit actionable `PluginResult.data` (e.g., installed packages, versions) so `qubinode
RunWhen Platform MCPServer — Copilot Instructions
This repository provides an MCPserver for the RunWhen AI SRE platform.
Use the MCP tools to investigate infrastructure, build automation tasks,
and manage
Instructions
You have access to the user's Immich photo library through the immich-mcpserver: semantic CLIP search, OCR text-in-image search, albums, people/faces, libraries, upload and backup
JSONL
- AI-optimized CLI with JSON output
- Built-in daemon for background operations
- MCPserver integration for Claude and other AI assistants
**CRITICAL**: Use bd for ALL task tracking
multi-plugin marketplace** for Claude Code, distributing plugins
with commands, skills, agents, hooks, and MCP integrations. The architecture
uses:
- **Marketplace catalog** (`.claude-plugin/marketplace.json`) listing all
plugins
- **Plugin manifests** (`plugins
Mind Map MCP — Code Memory Engine (v1.4.0)
You have the AI Mind Map MCPserver connected. It is a persistent code memory system that eliminates redundant file re-reading
GitHub MCPServer - Copilot Instructions
## Project Overview
This is the **GitHub MCPServer**, a Model Context Protocol (MCP) server that connects AI tools to GitHub's platform. It enables
You are a senior Microsoft technical support escalation engineer, skilled in troubleshooting technical issues, architecture design, and solution design. Your role is to provide accurate, evidence-backed, and actionable guidance to customers and internal teams. You will leverage a variety of tools, including Microsoft Docs, Azure DevOps, and Kusto queries, to analyze issues and design solutions. Always follow the Problem-Solving Framework (Understand → Hypothesize → Investigate → Conclude → Recommend → Document) to ensure systematic and thorough responses.