learn

Extract patterns, decisions, anti-patterns, and quality-rule candidates from campaign artifacts.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/mferris77/SpringBoard --skill learn-mferris77
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn
Source: https://github.com/mferris77/SpringBoard/tree/main/vscode-citadel-harness/skills/learn
Command: npx skills add https://github.com/mferris77/SpringBoard --skill learn-mferris77

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Campaigns accumulate decisions, telemetry, and informal lessons that are hard to surface and reuse. This Skill automates extraction of successful patterns, anti-patterns, key decisions, and candidate quality rules from completed campaign artifacts so teams retain institutional knowledge and reduce repeated mistakes.

Core Features & Use Cases

  • Extracts successful patterns and evidence from campaign files, postmortems, and audit telemetry to inform future work.
  • Identifies failed patterns and anti-patterns with avoidance guidance and links to evidence.
  • Captures key decisions and outcomes from Decision Logs or inferred from phase descriptions.
  • Proposes medium-or-higher confidence quality rule candidates and appends non-duplicative rules to .claude/harness.json.
  • Use case: post-campaign knowledge capture for engineering/harness teams to prevent regressions and automate code-quality enforcement.

Quick Start

Run the learn skill on a completed campaign to extract patterns, write .planning/knowledge files, and optionally append high-confidence quality rules to your harness.

Frequently Asked Questions about learn

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

FAQPage Schema
How do I extract reusable patterns from postmortems and campaign files?

Pattern extraction from postmortems processes completed campaign markdowns and audit.jsonl telemetry to identify successful patterns, anti-patterns, and key decisions, writing actionable knowledge to .planning/knowledge/.

How does campaign retrospective learning work with audit logs?

Campaign retrospective learning parses audit.jsonl telemetry and campaign artifacts to infer decisions and outcomes, automatically capturing institutional knowledge to prevent repeated mistakes across engineering harnesses.

Can I automatically generate quality rules from completed campaign telemetry?

Yes, you can generate quality rules from campaign telemetry by extracting evidence-backed, medium-or-higher confidence rule candidates and appending non-duplicative rules directly to .claude/harness.json for automated code-quality enforcement.

What is the best way to capture anti-patterns and avoidance guidance from engineering campaigns?

Capturing anti-patterns from engineering campaigns involves analyzing failed patterns in postmortems and audit logs, then generating knowledge files with avoidance guidance and evidence links for future reference.

Do I need completed campaign files and postmortems to extract knowledge?

Yes, completed campaign files, postmortems, and audit.jsonl telemetry are required inputs, as the extraction process relies on these artifacts to identify actionable patterns and decisions for your knowledge base.

When should I not use automated pattern extraction on campaign artifacts?

You should avoid automated pattern extraction when campaigns lack sufficient evidence or when proposed rules fall below the medium confidence threshold, as the system enforces evidence-backed requirements to prevent low-quality knowledge generation.