skills.rest
crycriM
crycriMCOMMUNITY
Explore all skills by this author →
GitHub

supertank

Ingest PDFs, URLs, and text into SQLite and ChromaDB with KeyBERT keywords.

Updated Apr 19, 2026
Search Tags:#pdf#semantic-search#knowledge-base#ingest#text-extraction#chromadb#keybert
One-click install
npx skills add https://github.com/crycriM/hermes-skills --skill supertank
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: supertank
Source: https://github.com/crycriM/hermes-skills/tree/main/data-science/supertank
Command: npx skills add https://github.com/crycriM/hermes-skills --skill supertank

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires keybert, pymupdf, pdfplumber, pytesseract, sentence-transformers, chromadb, beautifulsoup4, httpx, and includes scripts (resource) components.

What problem does it solve?

Ingest and organize research articles, PDFs, and posts into a scalable, searchable knowledge reservoir using SQLite indexing, KeyBERT keywords, and ChromaDB embeddings.

Core Features & Use Cases

  • Ingest multi-source content (PDFs, URLs, text) into a structured knowledge store.
  • Deduplicate content and merge keywords and references for consistent indexing.
  • Generate keywords, summaries, and semantic embeddings for fast retrieval; supports end-to-end querying via RAG.
  • Use cases include academic literature curation, research workflows, and finance/algo-trading knowledge management.

Quick Start

Run the ingest command against a URL or file to populate Supertank and enable fast semantic search.

Frequently Asked Questions about supertank

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

FAQPage Schema
How do I build a searchable knowledge base from research PDFs and web posts?▼

You can build a searchable knowledge base by ingesting research PDFs and posts, extracting keywords with KeyBERT, and indexing semantic embeddings in ChromaDB. This enables fast retrieval and RAG querying across diverse sources.

What is the best way to extract text and keywords from PDFs for semantic search?▼

The best way to extract text and keywords from PDFs for semantic search involves using PDF text extraction libraries alongside KeyBERT. The pipeline applies deterministic ingestion to generate keywords, summaries, and vector embeddings for fast retrieval.

Does ChromaDB work well for organizing academic literature and finance research?▼

ChromaDB works well for organizing academic literature and finance research by storing vector embeddings generated from ingested content. It supports fast semantic search and retrieval-augmented generation across multi-source documents.

How does RAG querying work with a ChromaDB knowledge reservoir?▼

RAG querying with a ChromaDB knowledge reservoir works by retrieving semantically similar indexed documents using vector embeddings. The system matches user queries against stored embeddings to provide context-aware summaries and fast retrieval.

Can I ingest and deduplicate content from multiple URLs and PDF files automatically?▼

You can ingest and deduplicate content from multiple URLs and PDF files automatically. The system deduplicates content while merging keywords and references to maintain consistent indexing across all sources.

Related Skills

Explore More Data & Analytics Skills→
eddieluengeddielueng
1

ocr-and-documents

Extract text from PDFs and scanned documents via URL or local OCR tools.

Community
Intermediate
robertbr123robertbr123

chroma

Manage and query vector embeddings with metadata using chromadb.

Community
Advanced
qcmuuqcmuu

sentence-transformers

Generate semantic embeddings from text using sentence-transformers models.

Community
Intermediate
adeboldadebold

AgentDB Vector Search

Perform vector-based semantic search against an AgentDB index.

Community
Advanced