Brad Zhang / public archive

Brad Zhang

AI product notes, agent workflow writing, and open-source dossiers for founders and early technical teams.

GitHubX

Open source/skill-knowledge-extractor

Project dossier / README

skill-knowledge-extractor

yizhiyanhua-ai/skill-knowledge-extractor

A tool for turning tacit workflow know-how into extractable assets.

The valuable part of AI work is often implicit. Extraction tools make that hidden layer easier to package and reuse.

Open GitHub

Repository reading order

Understand the workflow it replaces before deciding whether to stay in the README, open a topic path, or move toward collaboration.

01 / Workflow

Knowledge staying trapped inside long chats, docs, and operator memory.

Start by naming the old workflow this repo is replacing.

02 / README

Skill Knowledge Extractor 🇨🇳 中文 Extract reusable knowledge patterns from conversations, documents, or work experiences to create Skills.

Then use the README and site interpretation to confirm the project boundary and delivery surface.

03 / Proof graph

Return to thesis and the wider proof system

Finally, place the repository back into a bigger market question and working conversation.

Dossier links

Turn raw repository facts into a founder-readable evidence chain.

Raw repo

Return to GitHub for the raw repository, code, and issue state.

Proof library

Place this repository back into the broader public project library.

Topic cluster

Continue into the thesis, market question, and capability lane it supports.

github.com/yizhiyanhua-ai/skill-knowledge-extractor

Skill Knowledge Extractor

🇨🇳 中文

Extract reusable knowledge patterns from conversations, documents, or work experiences to create Skills.

Features

  • Analyze conversation records (sales calls, customer service chats)
  • Extract patterns from work documents and notes
  • Generate Skill drafts automatically
  • Support both Chinese and English

Usage

# Analyze a conversation file
python scripts/extract_patterns.py --input chat.txt --type conversation

# Analyze a document
python scripts/extract_patterns.py --input notes.md --type document

# Output as JSON
python scripts/extract_patterns.py --input data.txt --json

Pattern Types

  • Opening patterns - Standard greetings and introductions
  • Diagnosis patterns - Problem discovery questions
  • Objection handling - Responses to customer concerns
  • Closing patterns - Deal closing techniques
  • Checklist patterns - Verification items
  • Template patterns - Reusable document structures
  • Decision tree patterns - Conditional logic flows
  • Procedure patterns - Step-by-step workflows

License

MIT

Proof-system routes

This repository is not the endpoint. It should keep flowing into archive, forum, thesis, and collaboration.

Longform route

Return to the archive for the argument behind this build

Read the repository and the writing as one public proof chain.

Forum route

Reconnect the project to launches, field notes, and shorter experiments

The forum layer carries pace, live context, and iteration.

Collaboration route

Turn this project page into a concrete collaboration path

Founding engineer, design partner, and contract build paths converge here.

Related clusters

A project page should also return to the bigger market question.

01 / Cluster 01

3 routes

Harness engineering for agentic AI products

What has to surround the model before an AI workflow becomes a product?

02 / Cluster 02

3 routes

Long-running agents and resumable work

Where does the user lose trust when an agent runs for hours?

03 / Cluster 03

3 routes

Developer communication for AI infrastructure

Can someone understand the product before they understand the whole codebase?

Commercial outcomes

Founding engineer fit

Design partner sprint

Developer-facing proof work