borghei/Claude-Skills / research/patent
patent
VerifiedPatent research and IP landscape: prior-art search, claim mapping, landscape analysis, and freedom-to-operate. Use when conducting a prior-art search, mapping an IP landscape, evaluating patentability, or running an FTO analysis.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 8.4 KB |
| references/claim-mapping-and-landscape.md | markdown | 6.8 KB |
| references/freedom-to-operate-and-patentability.md | markdown | 8.0 KB |
| references/prior-art-search-strategy.md | markdown | 6.2 KB |
| scripts/claim_landscape_mapper.py | script | 7.4 KB |
| scripts/patentability_scorer.py | script | 10.1 KB |
| scripts/prior_art_search_planner.py | script | 7.0 KB |
references/freedom-to-operate-and-patentability.md
View on GitHub# Freedom-to-Operate & Patentability Reference
Reference for FTO analysis and patentability assessment.
## 1. FTO vs patentability — different questions
| Question | Patentability | FTO |
|----------|---------------|-----|
| Asks | Can I get a patent on my invention? | Can I make/use/sell my product without infringing? |
| About | Novelty + non-obviousness + utility | Active patents whose claims my product might practice |
| Driven by | Prior art (any date pre-filing) | Active patents only (typically 20 years from filing) |
| Outcome | Patent granted (or not) | Risk assessment + mitigations |
| Output | Patent claims | Risk register + legal opinion |
You can get a patent and still infringe someone else's patent. Both
analyses are necessary for different decisions.
## 2. FTO process
### Step 1: Identify candidate blocking patents
- Search patents in your product's technology area
- Focus on patents currently in force (not expired)
- Include patents owned by competitors + NPEs (patent assertion entities)
- Cover all jurisdictions you plan to sell in (IP is territorial)
### Step 2: Construe claims
- Read claims carefully
- Apply ordinary meaning to terms (or specification-defined meaning)
- Check prosecution history (file wrapper) for narrowing arguments
- Identify each claim element
### Step 3: Map to product
- Element-by-element comparison
- Does your product practice every element?
- Consider doctrine of equivalents
### Step 4: Risk classification
- **High risk:** likely infringement of valid claim
- **Medium risk:** some risk; design-around possible
- **Low risk:** product doesn't practice key elements
- **Cleared:** confident no infringement
### Step 5: Mitigations
- Design-around (modify product to avoid element)
- License the patent
- Challenge validity (IPR/PGR/litigation)
- Wait for expiration
- Abandon market segment
- Accept risk (legal opinion as defense)
### Step 6: Legal opinion
- Patent attorney issues formal opinion
- Provides defense against willful infringement (treble damages)
- Costs $5-30K depending on scope
## 3. The infringement test
For literal infringement, the accused product must practice **every element** of at least one claim.
If one element is missing, no literal infringement of that claim.
Doctrine of equivalents may still apply for substantially-similar elements.
## 4. Claim construction matters
How a claim is read affects infringement:
- **Plain meaning rule** — give terms their ordinary meaning
- **Specification informed** — if spec defines a term, use that
- **Prosecution history estoppel** — if inventor narrowed during prosecution, can't recapture in litigation
- **Doctrine of equivalents** — equivalent elements infringe
In litigation, claim construction (the "Markman hearing") is often outcome-determinative.
## 5. Patent validity challenges
If you face a problematic patent, consider:
### Inter Partes Review (IPR, US PTAB)
- Cost-effective challenge ($300K-$1M)
- Based on patents + printed publications only
- ~80% institution rate; ~60% of instituted claims canceled
### Post-Grant Review (PGR, US PTAB)
- Within 9 months of issue
- Broader grounds than IPR
- Less common
### Ex Parte Reexamination (US)
- Cheaper; less control
- Patent holder participates fully
### District court invalidity defense
- Counter-claim in infringement suit
- Most expensive route
### EPO opposition
- Within 9 months of grant
- Multi-party
- Cheaper than EU national invalidation
## 6. Patentability analysis
### Novelty (35 USC §102 / EPO Art. 54)
The invention must not be disclosed in any single prior art reference before the filing date.
Test: does ANY single reference disclose every element of the claim?
If yes → not novel; claim invalid / unpatentable.
### Non-obviousness / inventive step (§103 / Art. 56)
The invention must not be obvious to a person of ordinary skill in the art.
US Graham v. Deere factors:
- Scope and content of prior art
- Differences between claim and prior art
- Level of ordinary skill
- Secondary considerations (commercial success, long-felt need, unexpected results)
EU "problem-solution" approach:
- Identify closest prior art
- Determine differences and effects
- Define objective technical problem
- Would skilled person solve it this way (without invention)?
Most patents are rejected on §103 (obviousness), not §102 (novelty).
### Utility (§101 / industrial applicability)
- Must be useful (US: low bar; one specific use suffices)
- Must be capable of industrial application (EU)
### Subject matter eligibility (US §101 — major issue)
- Abstract ideas not patentable (Alice test)
- Mathematical algorithms alone not patentable
- Software claims must be tied to specific technical improvement
- Diagnostic methods often rejected
- "Significantly more" required beyond the abstract idea
### Enablement (§112)
- Description must enable skilled person to make + use the invention
- Without undue experimentation
- Genus claims need representative examples
### Definiteness (§112)
- Claim language must be clear enough that scope is determinable
- "Means-plus-function" claims have specific rules
## 7. Subject matter eligibility (US — tricky)
Two-step Alice/Mayo test:
### Step 1: Is the claim directed to a patent-ineligible concept?
- Abstract idea
- Law of nature
- Natural phenomenon
### Step 2: Does the claim include "significantly more"?
- Specific implementation
- Improvement to computer functioning
- Application that transforms the abstract idea
Software claims often fail Alice unless they:
- Improve computer technology specifically
- Solve a technical problem in a particular way
- Don't merely automate a known process
EU patent law is generally friendlier to software (technical character test).
## 8. Patentability scoring framework
A useful 5-dimension assessment:
| Dimension | Question | Pass criterion |
|-----------|----------|----------------|
| Novelty | Any single ref. discloses all elements? | No |
| Non-obvious | Obvious to skilled artisan? | No |
| Utility | Useful for a stated purpose? | Yes |
| Subject matter | Abstract idea / natural law? | No / "significantly more" |
| Enablement | Adequately described to enable? | Yes |
A patentable invention passes all five.
## 9. Defensive publication — the alternative
If filing isn't worth it (cost > value):
- Publish the invention publicly
- Becomes prior art against everyone (including you)
- Cheap (~$50 for a defensive publication on IP.com)
- Effective: prevents competitors from patenting it
Good for: ideas you want freedom to use but don't want to invest in patenting.
## 10. Trade secret as alternative
Sometimes trade secret is better than patent:
| Trade secret | Patent |
|--------------|--------|
| Indefinite duration | 20 years max |
| No filing cost | $30K-$100K+ |
| No public disclosure | Public after 18 months |
| Independent invention OK by others | Excludes all others |
| Reverse engineering = lost | Excludes even RE |
| Best for: processes, formulas | Best for: products, methods exposed in commercialization |
Coca-Cola formula: trade secret. Pharmaceutical compounds: patent.
## 11. Provisional patent application
- US-specific (1-year priority placeholder)
- Cheaper than utility (~$3-5K vs $20K+)
- No examination; no claims required
- 12 months to file utility application
- "Patent pending" status
- Good for: early-stage; trade-show disclosure; investor demos
Don't over-use provisionals: 60% of provisionals never convert. Strategic provisionals only.
## 12. Common pitfalls
- **Filing without prior-art search.** Examiner finds it; patent invalid.
- **Disclosing before filing.** Loses patentability outside US.
- **Confusing FTO with patentability.** Different analyses.
- **Skipping FTO before launch.** Surprise injunctions.
- **Ignoring NPE risk.** Patent assertion entities pursue successful products.
- **No legal opinion.** Treble damages on willful infringement.
- **Over-claiming.** Broad claims often invalidated; narrower may survive.
- **Forgetting continuations.** Maintaining pendency for strategic flexibility.
- **One-and-done landscape.** Patent landscapes shift constantly.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |