session-knowledge-harvest

Extract architectural knowledge from coding sessions and update documentation.

1|Updated Apr 14, 2026
One-click install
npx skills add https://github.com/bhanford9/vs-code-copilot-tools --skill session-knowledge-harvest
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-knowledge-harvest
Source: https://github.com/bhanford9/vs-code-copilot-tools/tree/main/skills/session-knowledge-harvest
Command: npx skills add https://github.com/bhanford9/vs-code-copilot-tools --skill session-knowledge-harvest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Extract and document architectural knowledge discovered during coding or investigation sessions. Mines the session for business rules, behavioral contracts, counterintuitive code patterns, new terminology, and gaps that were researched or user-clarified — then integrates the findings into the existing architecture knowledge base. Use at the end of any session where meaningful domain or codebase understanding was gained: debugging sessions, code reviews, work item analysis, investigation spikes, refactoring work, or any session where "I didn't know that before" occurred.

Core Features & Use Cases

  • Mine sessions for knowledge candidates and update the architecture documentation accordingly.
  • Check for gaps against existing architecture docs and plan targeted updates.
  • Support a Lessons Learned feedback loop via LessonsLearned.md and LessonsLearned.GLOBAL.md to improve documentation over time.

Quick Start

Run this skill at the end of any session where meaningful domain or codebase knowledge has been gained to extract and integrate new insights into the architecture docs.

Frequently Asked Questions about session-knowledge-harvest

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

FAQPage Schema
How do I extract architectural knowledge from a coding session?

Use this skill at the end of any debugging, code review, spike investigation, or refactoring session to mine newly discovered domain terms and rules, updating your architecture knowledge base and LessonsLearned.md files accordingly.

When should I document lessons learned from a debugging session?

Document lessons learned whenever a debugging session, code review, or investigation spike reveals new domain terminology, business rules, or counterintuitive code patterns that update your existing architecture knowledge base.

What is the best way to update architecture documentation after a code review?

Check for gaps against existing architecture docs, identify knowledge candidates mined from the code review session, and plan targeted updates to integrate newly discovered behavioral contracts and domain rules into the architecture knowledge base.

Can I integrate investigation spike findings into a global lessons learned file?

Yes, investigation spike findings can be integrated into LessonsLearned.GLOBAL.md by guiding newly discovered insights and domain rules into both local and global LessonsLearned.md files to support a continuous documentation feedback loop.

Does this knowledge harvesting approach work for refactoring work and code reviews?

Yes, this knowledge harvesting approach applies directly to refactoring work and code reviews, mining any session where new domain terms or behavioral rules are discovered and codifying them into the architecture knowledge base.