agentleFS
flowleap-aiflowleap-ai/flowleap-cli/skills/recipe-claim-drafting/SKILL.md

recipe-claim-drafting

Skill1 starsChanged 4 months agoMITView on GitHub

Prosecution recipe for drafting patent claims grounded in prior art — search the closest art, study its claim language, decompose the invention into elements, iterate drafts through claim analysis, and check formal drafting rules against MPEP/EPO guidelines. Trigger when the user asks to draft or improve patent claims, write independent and dependent claims, or stress-test draft claims against known art and formal requirements.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    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:

    Write the queries yourself, tuned precise — the goal is the closest few documents to read in full, so add a second discriminating ta term …

    line 19
  2. Distinctive:

    Write a one-paragraph invention summary, then decompose it yourself: list every element the invention needs to work (the mechanism, not the …

    line 38
  3. Very specific:

    Read the rules from the office's own page with flowleap-legal: EPC Art.

    line 57
  4. Distinctive:

    Check each claim against the fetched rules: antecedent basis, single-sentence form, no result-only language without structure, support in …

    line 62
  5. Distinctive:

    If the searches surface art reading on the draft, narrow the novel element and repeat.

    line 74

≈ “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

Recipe: Claim Drafting

Draft claims that are novel over the closest art and formally sound.

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. The steps stay the same.

Step 1: Find the Closest Art First

Claims drafted blind get rejected. Anchor on the art before writing:

Write the queries yourself, tuned precise — the goal is the closest few documents to read in full, so add a second discriminating ta term and probe the count down toward tens, not thousands (method in flowleap-patent):

flowleap --json tools run search_patents query='<self-written CQL>' range=1-1 details=false
flowleap --json patent search --query "<self-written CQL>" --limit 20
flowleap --json uspto search --query "<self-written ODP Lucene>" --limit 20   # title + metadata only

Pull the claims of the 3-5 closest hits and study their language:

flowleap --json ops claims <closest-patent>
flowleap --json ops abstract <closest-patent>

Step 2: Element Inventory of the Invention

Write a one-paragraph invention summary, then decompose it yourself: list every element the invention needs to work (the mechanism, not the benefit), keep the language concrete, and note keywords and synonyms per element (the decomposition method is in recipe-claim-analysis, Step 4).

Sort the elements: which are old (in the closest art), which are the point of novelty. The independent claim needs the old elements for operability plus at least one novel element for allowance.

Step 3: Draft the Claim Set

  • Independent claim: preamble (technical field) + transition (comprising) + minimum element set that is both operable and novel
  • Dependent claims: one fallback feature each, ordered by commercial value — these are the amendment reservoir for prosecution
  • Draft method + apparatus (and CRM where relevant) parallel sets

Step 4: Check Formal Drafting Rules

Read the rules from the office's own page with flowleap-legal: EPC Art. 84 and Guidelines F-IV (clarity, conciseness, support); 35 USC 112 and MPEP 2173 (definiteness) and MPEP 2181 (means-plus-function). Quote each rule with its URL and edition.

Check each claim against the fetched rules: antecedent basis, single-sentence form, no result-only language without structure, support in the description.

Step 5: Iterate Against the Art

For each draft independent claim, write a search from its novel element the same way as Step 1 — the novel element's phrase is the discriminating term:

flowleap --json patent search --query "<CQL written from the novel element>" --limit 10

If the searches surface art reading on the draft, narrow the novel element and repeat. Stop when the closest surfaced art clearly lacks an element.

Output

  • Closest-art table (reference, claim 1 language, what it lacks)
  • Element inventory split into old vs. novel
  • Full claim set (independent + dependents) with a support map into the description
  • Formal-rules checklist with legal citations per rule

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.