Extract atomic claims with provenance and confidence from source materials.
npx skills add https://github.com/Miosa-osa/canopy --skill reduce-miosa-osa
Please help me install this Agent Skill. Skill: reduce Source: https://github.com/Miosa-osa/canopy/tree/main/library/skills/processing/reduce Command: npx skills add https://github.com/Miosa-osa/canopy --skill reduce-miosa-osa
Transform raw input such as articles, transcripts, meeting notes, or documents into structured, atomic claims with provenance and confidence, reducing manual sifting.
Run /reduce on a transcript or document to generate atomic, provenance-backed claims.
High-intent search queries and answers about installing and using this skill.
To extract atomic claims from raw articles, transcripts, or documents, you can process Markdown, text, or PDFs to generate structured YAML outputs that include provenance metadata, confidence scores, and topic tags.
Yes, you can batch process multiple PDFs for knowledge extraction. The system supports batch processing options to systematically convert raw source materials into structured, provenance-backed atomic claims.
Tracking provenance when extracting insights from meeting transcripts is handled automatically by outputting YAML-structured claims that link each atomic insight back to its specific source location and context.
You classify and tag claims by topic and confidence during the extraction process itself, which outputs structured YAML data designed to reduce manual sifting and feed directly into downstream analysis workflows.
Yes, this extraction tool works with Markdown and plain text inputs, alongside PDF files. It ingests these formats to transform raw materials into structured, atomic claims with provenance metadata.
You should use automated claim extraction instead of manual sifting when processing large volumes of articles, transcripts, or documents for research and knowledge work, enabling structured outputs with confidence scores and provenance.
matteotittaProcess raw source material into structured, cross-referenced context files.
kinnerkarmanishExtract atomic claims with provenance, confidence, and topic tags from unstructured documents.
EurecaMomentExtract claims with supporting evidence from paper collections for AI benchmarks.
EnztunAnalyze video content and generate summaries using the z-ai-web-dev-sdk.
