generate-mcp-briefing

Discover MCP servers and tools, classify them by mode, and write the briefing file.

Updated Jan 31, 2026

Search Tags:#automation #mcp #classification #knowledge-base #tool-discovery #tool-briefing

Ready to run

npx skills add https://github.com/doesitscript/dotfile-vnext --skill generate-mcp-briefing

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: generate-mcp-briefing
Source: https://github.com/doesitscript/dotfile-vnext/tree/main/.cursor/skills/generate-mcp-briefing
Command: npx skills add https://github.com/doesitscript/dotfile-vnext --skill generate-mcp-briefing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Keeps MCP tool knowledge current by automatically discovering MCP servers and tools, classifying them by mode and trigger, and writing the persistent Tool Mode Map file to a known location for agents to consult at conversation start.

Core Features & Use Cases

Discover MCP servers and tools dynamically from live sources without hardcoding server names.
Classify tools by applicable modes (Agent, Plan, Debug, Ask) and determine their trigger conditions.
Generate or evaluate the MCP briefing file at .cursor/rules/02--cussorrules-mcp-briefieng-GENERATED.mdc to keep knowledge up to date and to surface drift.
Provide a mechanism to review uncertain placements before applying changes to the knowledge base.

Quick Start

Run the generate-mcp-briefing skill to refresh the MCP tool knowledge base after MCP changes.

Frequently Asked Questions about generate-mcp-briefing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I keep MCP tool knowledge up to date automatically after server changes?▼

To keep MCP tool knowledge up to date, run a briefing generation skill that dynamically discovers MCP servers and tools, classifies them by mode and trigger, and writes the persistent knowledge map to .cursor/rules for agents to consult at conversation start.

What is MCP tool classification by mode and trigger condition?▼

MCP tool classification by mode and trigger condition is the process of categorizing discovered tools into applicable operational modes like Agent, Plan, Debug, or Ask, and defining the specific conditions that should invoke each tool during an agent conversation.

How do I discover MCP servers and tools dynamically from a configuration file?▼

You can discover MCP servers and tools dynamically by reading the server names defined in the .cursor/mcp.json configuration file, which allows the generation process to query live sources without hardcoding server names.

Can I review uncertain MCP tool classifications before they are applied to the knowledge base?▼

Yes, you can review uncertain MCP tool classifications before they are applied, as the briefing generation process provides a mechanism to surface drift and uncertain placements for human review before writing changes to the persistent tool knowledge base.

Where should generated MCP briefing files be saved for Cursor agents?▼

Generated MCP briefing files should be saved to .cursor/rules/02--cussorrules-mcp-briefieng-GENERATED.mdc, ensuring agents consult this persistent Tool Mode Map at the start of every conversation to stay current with tool availability.

What is the best way to handle MCP tool drift in a Cursor workspace?▼

The best way to handle MCP tool drift is to run an automated briefing generator that compares live server states against the persistent knowledge base, surfaces the drift for review, and updates the .cursor/rules file with the latest tool classifications.