skill-awareness — resolving dependency trees, wiring skills into agent directories, and configuring MCPservers.
**Design principle: stay as close to npm as possible.** skillpm should feel like npm. Same `package.json
loop for anything irreversible; headless/scheduled runs are read/review-only by default; never assume an MCPserver exists; document the why. Full rules: .claude/CLAUDE.md §8.
- Specs precede code. Write small, safe, reversible
solution from internal knowledge; all methods require external verification.
### 2. Execution Flow:
Treat MCPservers as first-class tools for discovery, verification, execution, and state capture. PREFER CLI interactions (running
Keep `SKILL.md` description ≥ 50 chars, third-person, with specific trigger phrases.
- Use HTTPS/WSS for MCPserver URLs, not HTTP/WS.
### Emergency Recovery (3 Steps)
1. Re-run minimal local validation
packages that AI agents load at runtime to perform complex, multi-step tasks. The MCPserver makes skills discoverable and exposes their tools to any MCP-compatible client.
This
GitHub Copilot Custom Instructions — Agentic Helper
You have access to the `syzygy-mcp-layer` MCPserver. Use these tools on
every non-trivial coding task to maximize output quality
green. Never create a PR targeting `main` from any branch other than `dev`.
## Projektüberblick
MCP-Server für einen Raspberry Pi. Bietet MCP-Tools zur Steuerung von GPIO-LEDs (via `gpiozero
first contact), and
`.agent/skills/mcp-bridge/` (the same codebase-memory/session-memory
capabilities as a live MCPserver, for hosts that aren't Claude Code —
irrelevant if you're reading this file, since
Copilot Instructions
## MCPServer Development Guidelines
This is an MCP (Model Context Protocol) server project that provides enhanced context and capabilities to AI assistants.
### Key Features
- **Workspace Analysis**: Comprehensive analysis
GitHub Stars Contributions MCPServer - AI Coding Instructions
## Architecture Overview
This is a Model Context Protocol (MCP) server built with FastMCP that provides tools for managing GitHub Stars contributions
Aegis?
**MCP Aegis** is a Node.js testing framework designed specifically for **Model Context Protocol (MCP) servers**. It's the definitive tool for validating MCPserver implementations through both declarative YAML
AUDISTO-MCP: Codebase Guide
## Project Overview
AUDISTO-MCP is an MCPserver that connects AI agents to Audisto's technical SEO API (v2.0). It provides three tools: `get_help
existing L0 state instead of creating parallel coordination systems
## Hindsight Memory
Hindsight MCPserver is available at `http://localhost:8888/mcp/`. Use it deliberately.
**Session start** — before any non-trivial task
documentation, technical knowledge, or FAQ lookup,
use the `ask_tock` tool from the `tock-mcp-server`.
### Headers
Always include the `headers` argument when calling `ask_tock`.
| Header | Description | Values