← Back to Skills Marketplace
2770
Downloads
14
Stars
9
Active Installs
9
Versions
Install in OpenClaw
/install patent-scanner
Description
Describe your concept and discover what makes it distinctive — structured analysis for patent consultation. NOT legal advice.
Usage Guidance
This skill appears coherent and low-risk: it only asks for a description of your concept and produces local analysis and suggested claim framings. Before using it, consider: (1) do not paste highly sensitive source code or secret data into the prompt if you do not want that text stored or transmitted by your agent/environment; (2) the SKILL.md's claim that it 'operates entirely locally' is an instruction-level assertion — verify your agent/platform's data handling and network policies if you require strict local-only guarantees; (3) this tool is explicitly NOT legal advice — use it for technical ideation and then consult a qualified patent attorney for formal filings. If the skill's homepage or owner is unfamiliar to you, consider checking their reputation before sharing proprietary inventions.
Capability Analysis
Type: OpenClaw Skill
Name: patent-scanner
Version: 1.4.0
The OpenClaw AgentSkills skill bundle 'patent-scanner' is classified as benign. The `SKILL.md` file outlines a clear purpose for structured patent concept analysis and text generation. It explicitly states that the skill operates locally, does not transmit data, and does not modify, delete, or write any files. There are no instructions for the AI agent to perform any malicious actions, such as data exfiltration, unauthorized command execution, persistence mechanisms, or harmful prompt injection. External URLs (e.g., `https://obviouslynot.ai`) are only included as part of the output text for the user, not as instructions for the agent to interact with them. The skill includes appropriate disclaimers and warnings regarding sensitive data and legal advice.
Capability Assessment
Purpose & Capability
Name/description (patent concept analysis) match the SKILL.md: the skill asks for a concept description and provides a structured, multi-step analysis and claim-angle suggestions. It does not request unrelated binaries, credentials, or config access.
Instruction Scope
SKILL.md is an instruction-only workflow that stays within the stated purpose (component breakdown, combination analysis, scoring, and claim-angle generation). It does not instruct reading system files, environment variables, or contacting external endpoints. Note: the file explicitly states 'operates entirely locally' — this is a policy/assertion in the instructions rather than a technical guarantee; the agent/platform/environment ultimately controls whether data is transmitted.
Install Mechanism
No install spec or code files are present (instruction-only). Nothing will be written to disk by the skill itself, and no external URLs or packages are required.
Credentials
The skill requests no environment variables, credentials, or config paths. The level of access requested is proportional to an analysis-only helper.
Persistence & Privilege
always:false and user-invocable:true. The skill does not request persistent presence, system configuration changes, or access to other skills' credentials.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install patent-scanner - After installation, invoke the skill by name or use
/patent-scanner - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.4.0
Patent Scanner 1.4.0
- Removed the "disable-model-invocation: true" property, enabling model invocation.
- No other functional or content changes detected.
v1.3.0
Patent Scanner 1.3.0
- Output schema updated for consistency: renamed `id` to `pattern_id`, and `scores` to `score`.
- Minor clarification and formatting improvements throughout documentation.
- No changes to the core analysis framework or functionality.
v1.2.0
**Expanded patent-scanner with attorney-driven abstraction and value signals**
- Integrated patent attorney John Branch’s abstraction techniques to promote broader, implementation-agnostic pattern identification.
- Added "Problem-Solution-Benefit" mapping for each pattern, requiring specific articulation of technical limitation, mechanism, and measurable advantage.
- Introduced claim angle generation: for high-scoring patterns (8+), outputs method, system, and apparatus claim framings.
- Included new "Patent Value Signals" (market demand, competitive value, novelty confidence) as advisory context for prioritization.
- Updated help, tags, and clarifications throughout to reflect these methodology enhancements and a strictly local-data analysis process.
v1.1.3
SEC-1 through SEC-6: Safety statement fix, stealth-mode warning, model backend clarification, Sensitive Data Warning section, share_text warnings
v1.1.2
Security improvements: added safety boundaries, disable-model-invocation, privacy notes, updated homepages
v1.1.1
Migrated to public GitHub repo, updated homepage URLs
v1.1.0
- Added agent identity statement to clarify role, approach, and tone.
- Revised and condensed output formatting, adding clearer next steps and “share card” formats.
- Enhanced error handling and user guidance messages to be more concise.
- Expanded skill tags for improved discoverability.
- Updated share card and disclaimer instructions to ensure consistency and clarity.
- General copyediting for brevity, focus, and consistency throughout documentation.
v1.0.1
URL standardization, workflow text consistency, UPL compliance updates
v1.0.0
Initial release of Patent Scanner – analyze invention descriptions and identify potentially distinctive aspects.
- Provides structured scoring (distinctiveness, sophistication, system impact, frame shift) across up to 13 points.
- Breaks down inventions into components, analyzes combinations, and maps technical problem-solution fit.
- Outputs findings in clear markdown summaries and share-card formats, with scoring tables and synergy analysis.
- Includes actionable next steps (review, validate with validator skill, documentation, consult professional).
- Enforces strict terminology; not legal advice and always shows a required disclaimer.
Metadata
Frequently Asked Questions
What is Patent Scanner?
Describe your concept and discover what makes it distinctive — structured analysis for patent consultation. NOT legal advice. It is an AI Agent Skill for Claude Code / OpenClaw, with 2770 downloads so far.
How do I install Patent Scanner?
Run "/install patent-scanner" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Patent Scanner free?
Yes, Patent Scanner is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Patent Scanner support?
Patent Scanner is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Patent Scanner?
It is built and maintained by Lee Brown (@leegitw); the current version is v1.4.0.
More Skills