qmd-knowledge

Organize project learnings and issue notes into a centralized knowledge base via qmd MCP server.

114|13|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/jellydn/my-ai-tools --skill qmd-knowledge
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qmd-knowledge
Source: https://github.com/jellydn/my-ai-tools/tree/main/skills/qmd-knowledge
Command: npx skills add https://github.com/jellydn/my-ai-tools --skill qmd-knowledge

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires qmd, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a centralized, AI-assisted knowledge base for projects by capturing learnings, issue notes, and conventions using the qmd MCP server, reducing context switching and information loss.

Core Features & Use Cases

  • Unified repository for project learnings and issues, enabling quick lookup and historical context.
  • Automatic indexing and search through the qmd MCP server to retrieve relevant notes and decisions.
  • Cross-tool accessibility (OpenCode, Claude Code, Amp) so you can record and query knowledge from your preferred environment.
  • Use Case: During a project, record key learnings after important design decisions and instantly search for related issues and decisions when planning next steps.

Quick Start

Install the qmd tool and this skill, then create a project knowledge base by running the skill's recording commands to populate your ~/.ai-knowledges/<project-name>/ and index content with qmd embed. For routine use, record learnings with the script and query with qmd MCP to surface relevant context.

Frequently Asked Questions about qmd-knowledge

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

FAQPage Schema
How do I create a searchable knowledge base for project learnings and issues?

To create a searchable knowledge base for project learnings, you can use this Skill to automatically capture and store decisions and notes into a centralized repository, then run qmd embed to index the content for fast retrieval.

How does the qmd MCP server index project knowledge and conventions?

The qmd MCP server indexes project knowledge by enforcing a structured storage model at ~/.ai-knowledges and running qmd embed after changes, ensuring the searchable index of learnings and issues stays up to date.

Can I query project issue notes across different coding tools like Claude Code and OpenCode?

Yes, you can query project issue notes across different coding tools. This Skill operates across supported environments like Claude Code, OpenCode, and Amp to record and retrieve knowledge from your preferred tool.

Do I need the qmd tool installed to capture and search project decisions?

Yes, you need the qmd tool installed as a dependency. This Skill relies on the qmd MCP server to automate the organization, embedding, and querying of your project's historical context and learnings.

What is the best way to organize AI-assisted project learnings to prevent information loss?

The best way to organize AI-assisted project learnings is using a structured storage model at ~/.ai-knowledges, which automatically captures notes and conventions into a unified repository to prevent context switching and information loss.

How do I record key learnings after design decisions and search for related issues?

You record key learnings after design decisions by running the Skill's recording commands to populate your project directory, then use the qmd MCP to instantly search and surface relevant historical context and related issues.