drift-learn

Clusters weekly incident reports into compound lessons and skill patches.

Updated Jun 23, 2026
One-click install
npx skills add https://github.com/Walliiee/agent-harness --skill drift-learn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: drift-learn
Source: https://github.com/Walliiee/agent-harness/tree/main/skills/drift-learn
Command: npx skills add https://github.com/Walliiee/agent-harness --skill drift-learn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams make sense of recurring drift incidents by turning a week of incident records into durable lessons, so patterns do not get lost in individual postmortems.

Core Features & Use Cases

  • Incident clustering: Groups recent incidents by shared taxonomy and recurring failure modes.
  • Compound lesson generation: Synthesizes repeated root causes into actionable lessons and recommended skill patches.
  • Operational review support: Ideal for weekly reliability reviews where incident logs must be converted into a concise learning summary.
  • Use Case: A team with several memory-capture and cron-related incidents can run this Skill to identify the common pattern and append a structured lesson entry for the week.

Quick Start

Ask the agent to run the weekly drift learning pass on the latest incident records and summarize the recurring causes with recommended fixes.

Frequently Asked Questions about drift-learn

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

FAQPage Schema
How do I cluster weekly incident records into actionable postmortem lessons?

Clustering weekly incident records into postmortem lessons requires grouping drift incidents by shared taxonomy and recurring failure modes. This Skill processes dated incident markdown files to synthesize compound lessons and recommend skill patches for operational use.

What is the best way to synthesize recurring incident root causes for reliability reviews?

Synthesizing recurring incident root causes for reliability reviews involves analyzing recent incident ledgers to identify common failure patterns. This Skill converts individual postmortems into concise learning summaries, ensuring repeated reliability issues are captured.

Do I need specific markdown formats to analyze incident taxonomy and generate lessons?

Analyzing incident taxonomy requires reading dated incident markdown files as input. The Skill parses these records to cluster failure modes and appends structured lessons, making dated markdown formatting a prerequisite for downstream operational use.

Can I automate incident postmortem synthesis from recent incident ledgers?

Automating incident postmortem synthesis from recent incident ledgers is possible by running a weekly drift learning pass. This clusters the latest incidents, summarizes recurring causes, and appends structured lesson entries with recommended fixes.

How does clustering drift incidents improve postmortem analysis?

Clustering drift incidents improves postmortem analysis by grouping repeated failure modes into compound lessons rather than isolated events. This prevents patterns from being lost across individual postmortems and generates recommended skill patches for remediation.