learning-harvest

To capture and categorize learnings from completed tasks, analyze task outcomes to identify recurring issues and route insights to appropriate homes like AGENTS.md rules, tests, or wikis. This ensures valuable knowledge is retained and applied to future projects, reducing redundancy and improving task efficiency.

Retrospective analysis for project knowledge retention is the process of evaluating task outcomes to extract generalizable learnings and cost-saving opportunities. You need it after completing complex feature development or substantial tasks where identifying recurring issues can enhance existing skills and improve future task efficiency.

The best way to extract reusable learnings from a completed software development task is to run a learning harvest that analyzes outcomes and categorizes actionable items. Extracted best practices are then routed to wikis, tests, or AGENTS.md rules to sharpen existing skills or create new ones for similar future projects.

Yes, you can route categorized task learnings directly into AGENTS.md rules, tests, updated skills, or wikis. The learning categorization process evaluates task outcomes and directs actionable items to their appropriate homes, ensuring the extracted knowledge is formally applied to future project workflows.

This learning management approach works for any project workflow where knowledge retention and task efficiency are critical. It requires analyzing task outcomes and categorizing learnings into actionable items, making it applicable across diverse project environments without depending on specific platforms or external dependencies.