tech-distiller

Distill technical documents into structured Markdown reports filtered by direction.

144|17|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/yaofeino1/tech-distiller --skill tech-distiller
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tech-distiller
Source: https://github.com/yaofeino1/tech-distiller/tree/main
Command: npx skills add https://github.com/yaofeino1/tech-distiller --skill tech-distiller

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill turns scattered technical documents into a clear, structured technical report that’s easy for newcomers and reviewers to understand, while filtering content by role or technical direction.

Core Features & Use Cases

  • Structured technical distillation: Extracts architecture, algorithms, interfaces/contracts, performance metrics, design decisions, and open issues into a consistent report layout.
  • Direction/role-based filtering: Keeps only the parts relevant to a chosen direction such as 搜索, 推荐, 后端服务, 基础架构, 数据平台, ML/AI, 客户端, or 前端.
  • Report evolution workflows: Supports incremental updates by merging newly provided documents, with correction notes and version snapshots.

Quick Start

Tell the assistant: “帮我蒸馏这些技术文档,重点生成搜索方向的技术报告,面向新员工,输出为 Markdown。”

Frequently Asked Questions about tech-distiller

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

FAQPage Schema
How do I distill scattered technical documents into a structured architecture report?

To distill technical documents into an architecture report, the system fetches mixed sources like URLs and local files, extracts technical signals across architecture and algorithms, and generates a structured Markdown report for a specified audience.

Can I filter technical document summarization by a specific direction like search systems or ML?

Yes, technical document summarization supports direction-based filtering, keeping only content relevant to chosen fields like search systems, ML/AI, backend services, or data platforms for the final structured report.

What is the best way to generate an onboarding summary from internal knowledge-base entries?

Generating an onboarding summary involves ingesting knowledge-base entries, filtering content by role or direction, extracting design decisions and performance metrics, and outputting a consistent Markdown document tailored for newcomers.

How do I incrementally update a technical report with newly provided local files?

You can incrementally update a technical report by merging newly provided local files into existing content, applying correction notes, and generating version snapshots to maintain document evolution.

Does the technical distillation process support exporting reports to docx or pptx formats?

Yes, the technical distillation process supports exporting the final structured report to docx or pptx formats, alongside the default Markdown output, for versatile review and presentation needs.