Claude, Fto Triage

Can Claude clear a new product of patent risk before launch? Not quite, and the gap between what it sounds like it does and what it actually does is the important part. The fto-triage skill screens a described product or feature against a fixed set of claim elements, keywords, and jurisdictions, then returns a structured triage report. It does not clear anything.

What the skill actually screens against

A skill is a folder Claude reads before it works. This one is called fto-triage, and its SKILL.md file holds the full instruction set in plain language, with no hidden logic. Claude reads the file, then acts on it; the file is the program defining what happens. The specific constraint it runs is narrow: screen the described product or feature against the claim elements, keywords, and jurisdictions the file specifies, then flag anything that matches for a closer look.

How the pipeline runs

The instructions sit in a step section, and Claude reads each step in order, then runs it. The process is linear, with no branching unless a specific step calls for one. That means the same checklist produces the same shape of report, run after run, regardless of which product description gets fed into it.

What "flagged" actually means

Staying inside that checklist means the report holds its shape every time: same checklist in, same triage report out. A flag on the report means the description matched something in the claim elements, keywords, or jurisdictions the skill was told to screen against. It is a signal to look closer, not a legal conclusion about infringement, and the absence of a flag is not a clearance either. Deciding whether a product is actually clear to operate remains a legal judgment call that stays with the person running the skill, ordinarily made with patent counsel.

Running it on your own feature

The practical exercise is to give Claude a short description of a feature about to ship, ask it to read the fto-triage skill, and have it explain exactly what it needs from you before running any check. Then it triages the description against its checklist and explains what it flags and why. That sequence, stating its own inputs before acting, is the same discipline seen across this whole family of skills: fixed process, transparent about what it needs, silent on anything outside its checklist.

Why this distinction matters

Treating a triage report as a legal clearance would be a mistake with real consequences. The skill's value is in doing a consistent, repeatable first pass against a known set of claim elements and jurisdictions, catching what a fixed checklist can catch. It cannot substitute for the judgment of a patent attorney evaluating the full landscape of relevant patents, claim scope, and jurisdictional nuance that a checklist alone cannot capture.

Key takeaways

• The fto-triage skill screens a product description against claim elements, keywords, and jurisdictions specified in a fixed file; it does not clear a product of patent risk.
• The process is linear: read the file, execute each step, return the triage report, with no branching unless specified.
• A flag on the report means a match was found against the checklist, not a legal determination of infringement.
• No flags does not mean the product is legally clear to operate.
• The actual freedom-to-operate decision stays with the user and ordinarily requires patent counsel.

Who this is for

Product and legal teams who want a fast, repeatable first pass at screening a new feature against known patent claim elements before escalating to formal legal review, with a clear understanding that the tool triages rather than clears.