Turn single research papers into Obsidian notes with DeepPaperNote
Run a source-grounded agent workflow that reads one paper deeply and saves a durable Obsidian-style research note.
npx skills add agentskillexchange/skills --skill turn-single-research-papers-into-obsidian-notes-with-deeppapernote
Use DeepPaperNote when an agent needs to turn one paper into a structured, evidence-backed Markdown note that preserves the research question, methods, results, figure context, limitations, metadata, and reusable links for an Obsidian-style academic wiki. The operator gives a title, DOI, URL, arXiv ID, Zotero item, or local PDF; the skill resolves the paper, extracts source text and figures, runs quality gates, and saves the final note to the workspace or configured Obsidian vault. Invoke it instead of using a note app or PDF reader normally when the user wants an AI agent to perform the repeatable single-paper ingestion workflow end to end, with fail-closed behavior when evidence is insufficient. The scope boundary is the narrow one-paper deep-reading and note-production pipeline; it is not a general paper search engine, not a Zotero replacement, not an Obsidian product card, and not a generic PDF summarizer.
What this skill actually does
Inputs and prerequisites: One supported agent runtime such as Claude Code, Codex, Cursor, Copilot, or Gemini CLI; Python 3.10+; PyMuPDF; optional Zotero Local API, Semantic Scholar API, OCR tooling, and an Obsidian vault.
Setup notes: Run npx skills add 917Dhj/DeepPaperNote, choose the deeppapernote skill, install the PDF dependency with python3 -m pip install PyMuPDF, then provide a paper title, DOI, URL, arXiv ID, Zotero item, or local PDF and ask the agent to generate a deep-reading note.
Source and verification boundary: use https://917dhj.github.io/DeepPaperNote/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.