distill

Condenses prior LLM outputs or conversations into bullets, a paragraph, or a topic filter.

14|6|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/tyevans/tackline --skill distill-tyevans
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: distill
Source: https://github.com/tyevans/tackline/tree/main/skills/core/distill
Command: npx skills add https://github.com/tyevans/tackline --skill distill-tyevans

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill tackles information overload by reducing lengthy outputs to their core essence, making complex information digestible and actionable.

Core Features & Use Cases

  • Summarization: Condenses information into a specified number of bullet points or a single paragraph.
  • Topic Filtering: Extracts and condenses information relevant to a specific topic.
  • Use Case: After a lengthy /gather command that returns dozens of findings, use /distill to 5 bullets to get a quick overview of the most critical points.

Quick Start

Use the distill skill to condense the previous output into 3 bullet points.

Frequently Asked Questions about distill

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

FAQPage Schema
How do I condense verbose LLM output into a short summary?

You can condense verbose LLM output into a short summary by reducing lengthy text into a specified number of bullet points or a single paragraph, making complex information digestible and actionable.

Can I extract and condense conversation context to a specific number of bullets?

Yes, you can extract and condense conversation context into a specific number of bullets by specifying the desired count, which filters the most critical points from prior outputs into a quick overview.

How do I filter lengthy output for information relevant to a specific topic?

To filter lengthy output for a specific topic, apply topic filtering to extract and condense only the information relevant to that subject, prioritizing confirmed findings over likely or possible ones.

What is the best way to reduce information overload from lengthy conversation findings?

The best way to reduce information overload from lengthy conversation findings is to condense the verbose input into its core essence, prioritizing confirmed findings to make complex information digestible.

Does the distill skill work with raw conversation context or only piped output?

The distill skill works with both pipe-formatted upstream output and raw conversation context, handling either input type to reduce verbose text into essential points.