Convert articles or Markdown/TXT files into 5-8 structured knowledge cards for review and self-testing.
记忆
knowledge-base-publisher
试用Organize, format, and publish knowledge-base articles and documentation. Use when you need to convert raw notes, meeting transcripts, or scattered content into structured, publication-ready knowledge base entries with proper metadata, cross-references, and version tracking.
它能做什么
Organize, format, and publish knowledge-base articles and documentation. Use when you need to convert raw notes, meeting transcripts, or scattered content into structured, publication-ready knowledge base entries with proper metadata, cross-references, and version tracking.
技能文档
Knowledge Base Publisher
A skill for turning raw content into structured, publication-ready knowledge base articles.
When to Use
- Converting meeting notes into structured documentation
- Organizing scattered content into a knowledge base
- Adding metadata, tags, and cross-references to documentation
- Batch-processing multiple articles for publication
- Versioning knowledge base entries with changelog tracking
Prerequisites
- A source directory containing raw content (markdown, text, or transcript files)
- A target directory or platform for published output
- Optional: a taxonomy file for consistent tagging
Basic Steps
1. Ingest Raw Content
Read source files from the input directory. Supports:
.md/.markdownfiles.txtplain text files- Transcript formats (JSON with speaker/timestamp)
2. Structure & Enrich
For each document:
- Extract or generate title — from H1, filename, or AI summary
- Add metadata block — date, author, tags, related articles
- Normalize formatting — consistent headings, code blocks, lists
- Add cross-references — link to related articles by tag/topic matching
- Generate summary — one-paragraph abstract if none exists
3. Quality Gate
Check each article for:
- Minimum length (recommend ≥ 200 words)
- Has a clear title and at least one section
- No broken internal links
- Tags match the taxonomy (if provided)
4. Publish
Write structured articles to the output directory with:
- Consistent filename slugification
index.jsoncatalog with all article metadataCHANGELOG.mdtracking version history- Per-article
meta.jsonfor machine-readable metadata
Example Output Structure
knowledge-base/
├── index.json # Full catalog of all articles
├── CHANGELOG.md # Version history
├── articles/
│ ├── getting-started.md
│ ├── getting-started.meta.json
│ ├── api-reference.md
│ ├── api-reference.meta.json
│ └── ...
└── taxonomy.json # Tag/category definitions
Taxonomy Format (Optional)
{
"categories": ["tutorial", "reference", "guide", "faq"],
"tags": ["api", "setup", "configuration", "troubleshooting"],
"related": {
"api": ["getting-started", "authentication"],
"setup": ["configuration"]
}
}
Integration with TaskFlow
This skill works well as a TaskFlow child task:
- Parent flow defines the source directory and taxonomy
- Spawn this skill as a detached child task for batch processing
- Flow waits on child completion, then validates the output catalog
- Flow finishes with a summary of articles published
Integration with ClawHub
Published knowledge bases can be packaged as ClawHub skills:
- Structure the knowledge base as a skill folder with SKILL.md
- Use
clawhub publishto register it in the ClawHub registry - Others can discover and install via
clawhub search+clawhub install
相关技能
通过 WeKnora REST API 导入文档,并对知识库执行向量+关键词混合检索。
Answer Research KB questions from the team Gitea-backed knowledge base with OpenClaw reasoning, optional reference attachments, stable citations, OpenClaw-led evidence selection, and optional high-value Q&A persistence.
Import documents and retrieve knowledge through the Keystone REST API. Use for uploading files, URLs, or Markdown to a knowledge base; hybrid search within a...
Create, read, validate, and manage OKF (Open Knowledge Format) knowledge bundles for AI agents. Google's vendor-neutral markdown+YAML spec.
Produce a focused literature/knowledge review using only selected Research KB contents.