agentleFS
flowleap-aiflowleap-ai/flowleap-cli/skills/flowleap-npl/SKILL.md

flowleap-npl

Skill1 starsChanged 4 months agoMITView on GitHub

Search non-patent literature (scholarly works via OpenAlex) with year, open-access and publication-type filters, and resolve a cited paper (DOI, OpenAlex id or an office's citation string) to its record. Trigger when an agent needs journal articles, conference papers, theses, standards or preprints as prior art or scientific background, or needs to identify a paper an examiner cited — the first NPL source; flowleap-academic (Semantic Scholar/arXiv) is the second pass.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No patent-data key is needed, so this corpus stays live while EPO or USPTO is key-gated.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

FlowLeap NPL Search (OpenAlex)

Auth and global flags: see flowleap-shared.

In a chat client with the FlowLeap connector or in the FlowLeap Patent AI app, call the tools named in the flowleap-shared tool table instead of these commands, and never run a flowleap command there.

flowleap --json npl "perovskite solar cell stability" --limit 5
flowleap --json npl "CRISPR delivery" --from-year 2020 --to-year 2024 --open-access
flowleap --json npl "transformer attention" --type preprint
flowleap --json npl "hydraulic accumulator bypass" --to-year 2022 --type dissertation

Flags: --limit N, --page N, --from-year/--to-year, --open-access, --type journal-article|proceedings-article|book-chapter|preprint|review|book|dissertation|report|standard|dataset.

Order for an NPL sweep: npl first (OpenAlex, 250M+ works), then academic search (Semantic Scholar + arXiv) as a second pass for preprints and CS/physics work, deduplicated by DOI or title; a web search for what neither reaches (product manuals, web disclosures, standards bodies' sites). A year filter is coarse: check each publicationDate against the exact critical date.

Results include DOI, abstract, citedByCount, open-access URLs and author lists — use DOI + title when citing prior art.

Tools-facade equivalent: search_npl (query=, limit=, page=, and a filter object taking from_year, to_year, open_access, type).

Resolve a cited paper

Papers an office already cited are resolved, not searched:

flowleap --json npl-work 10.1038/nature14539
flowleap --json npl-work "SMITH J: 'Hydraulic actuator control', IEEE TRANS., 2015"
flowleap --json ops biblio EP3000000A3 --resolve-npl   # every cited NPL entry at once

npl-work takes a DOI, an OpenAlex work id (W…) or a citation string exactly as printed (a citedReferences[].npl entry of ops biblio, an NPL row of patent examiner-baseline, an isNPL cited document of citation search). Lookup order: OpenAlex id, then any DOI in the string, then the quoted title. status decides what you may say:

  • matched — work is the cited record (title, authors, venue, date, DOI, abstract, open-access link).
  • candidates — similar works, none certain. Say so; never pick one as the cited paper.
  • not_found — standards, product manuals and web pages are often absent from OpenAlex; cite the office's string as printed.

ops biblio --resolve-npl attaches the same result as nplWork to each citedReferences[].npl entry (one OpenAlex read per reference). For an EP family the citations sit on the A3 search-report publication.

Tools-facade equivalent: get_npl_work (reference=), and get_bibliography with resolve_npl=true.

No patent-data key is needed, so this corpus stays live while EPO or USPTO is key-gated. Papers are prior art in their own right, but they are a different corpus from patents — offer them as different data, never as a substitute for a gated office's search (see flowleap-keys).

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.