note-weaver-skill

To consolidate scattered notes into structured Markdown, the system extracts key claims from multi-source inputs like chat exports and voice transcripts, reorganizing them into logical sections such as background, core, and conclusion while removing semantic duplicates.

Semantic deduplication in note-taking identifies and merges conceptually overlapping text fragments while retaining distinct original viewpoints, ensuring that unique perspectives are preserved even when redundant information is removed during the consolidation process.

Generating a JSON-compatible knowledge graph from text fragments involves detecting concepts and entities within the notes, inferring causal or hierarchical relations, and outputting a structured JSON summary compatible with tools like Obsidian Graph View.

Yes, multi-source note fragments can be classified using both PARA and Zettelkasten patterns simultaneously, allowing the system to categorize notes by project areas while also creating atomic, linkable concepts for a connected knowledge base.

Applying scenario-specific note templates for book, meeting, and project tasks involves matching the raw input content to a predefined template structure, which then formats the fragmented text into standardized sections with appropriate metadata and tags.

The best way to automate periodic reviews from fragmented notes is to aggregate the structured Markdown entries over a timeframe, generating a summary that combines quantitative metrics with qualitative recommendations based on the consolidated knowledge.