knowledge-construction

Generate bilingual knowledge.json and summary.md from staged DocMason evidence.

135|36|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/JetXu-LLM/DocMason --skill knowledge-construction
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: knowledge-construction
Source: https://github.com/JetXu-LLM/DocMason/tree/main/skills/canonical/knowledge-construction
Command: npx skills add https://github.com/JetXu-LLM/DocMason --skill knowledge-construction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps unify and publish bilingual knowledge objects from staged DocMason evidence, ensuring traceability and structured outputs.

Core Features & Use Cases

  • Write bilingual Phase 3 knowledge objects (knowledge.json, summary.md) from the latest staged sources.
  • Generate sidecars and structured artifacts that preserve provenance.
  • Support governance workflows like knowledge-base-sync to ensure consistent, auditable results.

Quick Start

Run this workflow to translate staged evidence into bilingual knowledge.json and summary.md for the current bounded scope.

Frequently Asked Questions about knowledge-construction

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

FAQPage Schema
How do I generate bilingual knowledge objects from staged evidence sources?

Bilingual knowledge objects are generated by extracting structured data from staged DocMason sources to produce a knowledge.json file and a summary.md file that preserve source provenance and reference original artifacts.

How does provenance preservation work when publishing knowledge.json files?

Provenance preservation works by generating knowledge.json files and summary.md sidecars that explicitly reference source artifacts and derived affordances, ensuring every published knowledge object remains traceable to its original staged evidence.

What is the best way to automate bilingual summary generation for a knowledge base?

Automating bilingual summary generation is best achieved by targeting the governed packet and pending-work scope defined by knowledge-base-sync, reading pending_work.json manifests, and outputting structured summary.md files for each assigned source.

Do I need local file access and shell execution to build knowledge objects from DocMason evidence?

Yes, building knowledge objects from DocMason evidence requires local file access and shell execution to read structured manifests like pending_work.json, artifact_index.json, and derived affordances to ensure accurate and traceable results.

Can I use this workflow to process sources outside the knowledge-base-sync governed scope?

No, the workflow specifically targets the governed packet and pending-work scope defined by knowledge-base-sync, meaning it processes only assigned sources within that bounded scope to ensure consistent and auditable results.

What structured manifests are required to extract knowledge from staged sources?

Required structured manifests include knowledge_base/staging/pending_work.json, knowledge_base/staging/sources/*/knowledge.json, artifact_index.json, and derived affordances, which are all read to ensure accurate and traceable knowledge extraction.