agentleFS
luffysolution-svgluffysolution-svg/omnischolar/skills/paper-reading/SKILL.md

paper-reading

Skill4 starsChanged 2 months agoMITView on GitHub

Select a local Zotero PDF and publish its MinerU Markdown and assets for downstream reading. Use when a paper has not yet been parsed or the user explicitly requests a fresh parse; hand interpretation to literature-reading.

20 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When parsing is needed, call omnischolar_parse directly after the user has authorized the selected PDF's external upload.

    line 15
  2. Very specific:

    Do not set force unless the user explicitly requests a fresh parse and understands that it may upload the PDF again.

    line 16
  3. Distinctive:

    Read references/sync.md only for an explicit synchronization, repair, exclusion, conflict, or recovery request.

    line 20
  4. Distinctive:

    If MinerU is unavailable, Zotero indexed text may answer a narrow question when coverage is sufficient and the limitation is explicit.

    line 22

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Prepare a paper for local reading

This Skill owns attachment selection and MinerU publication, not interpretation. Its successful handoff is an exact publication.markdownPath, paper identity, and parser limitations.

Workflow

  1. Resolve the bibliographic parent and attachments with zotero_item in aggregate mode. If more than one PDF is available, list attachment keys and filenames, ask the user to choose, and do not parse or synchronize until an explicit attachmentKey is confirmed.
  2. Reuse an already known, readable publication.markdownPath unless the user requested a fresh parse or the source identity/version is inconsistent. Do not query or upload again merely because interpretation was requested.
  3. When parsing is needed, call omnischolar_parse directly after the user has authorized the selected PDF's external upload. The parse call performs MinerU cache lookup or upload before publication safety checks; do not make omnischolar_sync action=plan a mandatory parse preflight.
  4. Do not set force unless the user explicitly requests a fresh parse and understands that it may upload the PDF again. Never use it to bypass a conflict, exclusion, missing publication, or interrupted transaction.
  5. From the result, verify publication.markdownPath, bibliographic frontmatter, metadata.json, and sibling assets. Parsed image filenames begin with the Zotero key. Report OCR, formula, table, image, or layout limitations that affect downstream reading.
  6. Hand all interpretation, evidence location, and figure/table/equation reading to literature-reading. Do not write an interpretation file from this Skill.

Read references/sync.md only for an explicit synchronization, repair, exclusion, conflict, or recovery request. Those states govern publication, not whether MinerU may probe its cache or process an explicitly authorized PDF first.

If MinerU is unavailable, Zotero indexed text may answer a narrow question when coverage is sufficient and the limitation is explicit. Do not claim a complete paper interpretation from metadata, notes, annotations, or incomplete indexed text.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.