Knowledge Extraction Python Packages

deepsearch-toolkit

Interact with the Deep Search platform for new knowledge explorations and discoveries

openodke

openodke — an open implementation of ODKE+ (Apple, arXiv:2509.04696): ontology-guided knowledge extraction into a grounded knowledge graph. Independent; not affiliated with Apple.

ferret-mcp

MCP server that extracts complete knowledge from any codebase — architecture, patterns, dependencies, API surface

refactor-llm-analyzer

A new package designed to facilitate structured and reliable analysis of user input related to software refactoring in the context of LLM capabilities. It accepts a user's discussion or question about

take-minutes

Distill any conversation into structured knowledge. Local-first CLI for extracting decisions, ideas, questions, and action items from AI sessions, meetings, and chat logs.

yt-quick-insights

Python webapp for rapid extraction and analysis of YouTube content, enabling users to gain insights from videos and playlists without spending hours watching them.

id-svo-extractor

id-svo-extractor: Extract SVO triples from Indonesian text.

philosim

A new package that leverages the power of llmatch-messages to provide structured insights into complex philosophical and scientific topics, such as the simulation hypothesis. Users can input a specifi

text2structured-summary

text2structured-summary generates structured summaries from unstructured text using an LLM.

summarix

A new package designed to process user-inputted text statements or stories and extract structured summaries or insights using a reliable language model with pattern matching and retries. It simplifies

video-text-summarizer

A new package designed to analyze and summarize video content by processing pre-extracted textual information such as transcripts, subtitles, and descriptions. It leverages structured pattern matching