skills.rest
opensquilla
opensquillaOFFICIAL
Explore all skills by this author →
WebsiteGitHub

meta-knowledge-base-bootstrap

Bootstraps a domain knowledge base by classifying a seed source and indexing search results.

6.8k|540|Updated May 6, 2026
Search Tags:#ingestion#classification#XLSX#Search#Memory#Bootstrap#Knowledge Base
One-click install
npx skills add https://github.com/opensquilla/opensquilla --skill meta-knowledge-base-bootstrap
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: meta-knowledge-base-bootstrap
Source: https://github.com/opensquilla/opensquilla/tree/main/src/opensquilla/skills/exp/meta-knowledge-base-bootstrap
Command: npx skills add https://github.com/opensquilla/opensquilla --skill meta-knowledge-base-bootstrap

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Starting a new domain knowledge base from scratch requires manually researching a topic, saving findings, and organizing them into an index. This Skill automates that cold-start process in a single turn from one seed input.

Core Features & Use Cases

  • Source Classification: An LLM classifier labels the seed input as URL, PDF, GIT, or free-text topic.
  • Automated Ingestion: Runs the multi-search-engine skill (DuckDuckGo search) to gather relevant results as JSON.
  • Persistence and Indexing: Appends an ingestion summary to memory via memory_save and generates a kb-index.xlsx workbook with Engine, Title, URL, and Snippet columns.
  • Use Case: A researcher types "bootstrap a knowledge base on retrieval-augmented generation" and receives a memory entry plus a spreadsheet index of relevant sources in one turn.

Quick Start

Ask the agent to bootstrap a knowledge base from your topic or link, for example by saying "搭建知识库" or "bootstrap kb" followed by your seed topic or URL.

Frequently Asked Questions about meta-knowledge-base-bootstrap

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

FAQPage Schema
How do I bootstrap a knowledge base from a topic or URL?▼

Provide a single seed input such as a topic, URL, PDF path, or git repository and trigger the skill with phrases like "bootstrap kb" or "搭建知识库". The pipeline classifies the input, searches via multi-search-engine, saves a memory summary, and writes a kb-index.xlsx file.

What source types does the knowledge base bootstrap classifier support?▼

The classifier labels each seed as one of four types: URL for web links, PDF for .pdf paths or links, GIT for GitHub/GitLab or local repos, and TEXT for free-text topics. Currently the classification is informational and all types are ingested through the multi-search-engine skill.

Does the bootstrap skill route PDFs and git repos to specialized tools?▼

Not currently. An earlier design routed PDF seeds to pdf-toolkit and GIT seeds to a github skill, but those branches were dropped when the DSL moved to skill_exec. All inputs are ingested through multi-search-engine until per-classification routing is reintroduced.

What output files does the knowledge base bootstrap produce?▼

It appends an ingestion summary to memory/kb-bootstrap.md via the memory_save tool and creates a kb-index.xlsx workbook. The workbook contains columns for Engine, Title, URL, and Snippet populated from the search results JSON.

What happens if the meta-flow pipeline fails?▼

The documented fallback is to run the classifier prompt manually, invoke the appropriate ingestion skill directly, save the result with memory_save, and then build the xlsx index yourself using openpyxl.

Related Skills

Explore More Education & Research Skills→
OzeroHAXOzeroHAX
32

task-use-research-web-strategy

Research external web sources to compare technologies and find solutions.

Community
Intermediate
grandameniumgrandamenium
88

knowledge-base

Query an indexed organizational knowledge base before external research.

Community
Intermediate
surebelisurebeli

wiki-search

Search structured wiki pages by keywords, tags, and content relevance.

Community
Intermediate
Eng0AIEng0AI
15

langchain-retrieval

Index and retrieve document Q&A data using Supabase pgvector in LangChain pipelines.

Official
Intermediate