learnings-distill

Distill LEARNINGS.md into applied directives for HANDOFF.md planning.

Updated Jun 4, 2026
One-click install
npx skills add https://github.com/Kohlex/autopilot-plugin --skill learnings-distill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learnings-distill
Source: https://github.com/Kohlex/autopilot-plugin/tree/main/skills/learnings-distill
Command: npx skills add https://github.com/Kohlex/autopilot-plugin --skill learnings-distill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Distill learnings from LEARNINGS.md into a concise set of applied directives and drop the raw corpus at the session hop to prevent context-rot in execution.

Core Features & Use Cases

  • Distill phase-relevant items from LEARNINGS.md into actionable directives for the HANDOFF.md.
  • Enforce the drop boundary by ensuring the LEARNINGS.md path is not listed in the execution manifest.
  • Validate end-to-end gating with distillThenDropDecision to produce a PROCEED verdict or clearly documented findings.

Quick Start

Use the distill tool to convert LEARNINGS.md into applied directives and drop the corpus at the hop.

Frequently Asked Questions about learnings-distill

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

FAQPage Schema
How do I distill learnings from a markdown file into actionable planning directives?

To distill learnings, you convert raw LEARNINGS.md content into targeted applied-directive lines for planning, discarding the raw corpus at the session hop to prevent context-rot in execution.

What is the best way to prevent context-rot when handing off planning data to an execution session?

Preventing context-rot requires enforcing a drop boundary so the execution session reads only distilled guidance and foundation data, ensuring raw learnings paths are eliminated from the manifest.

How do I validate planning-context isolation before proceeding with phase execution?

You validate planning-context isolation using a gating decision mechanism that checks the distillation and drop boundary to produce a PROCEED verdict or clearly documented findings.

When do I need to drop the raw learnings corpus during per-phase planning cycles?

You need to drop the raw learnings corpus during per-phase planning cycles when you want to isolate reliable planning context and manage artifacts by passing only distilled guidance to execution.

Does distilling learnings into a handoff document remove manifest references?

Yes, distilling learnings into the handoff document removes manifest references by validating end-to-end gating and ensuring the raw learnings file path is not listed in the execution manifest.