legal-patent-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external patent data to generate legal reports and application drafts, creating a vector for instructions embedded in source data to influence output.
  • Ingestion points: Patent search results and bibliographic data are ingested via the moai-lawyer:legal-patent-search tool integration described in the Step 2 workflow.
  • Boundary markers: The skill does not explicitly define delimiters or specialized instructions to prevent the agent from following directives potentially contained within the retrieved patent text.
  • Capability inventory: The skill is capable of generating structured reports, FTO analysis, and patent claim drafts. It relies on internal analytical tools for these operations.
  • Sanitization: The workflow includes a post-processing chain involving 'moai-coworker:ai-slop-reviewer' and 'moai-writer:korean-humanize' to review and refine the generated text, which serves as a quality and safety filter.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:24 PM