jurisdiction-format

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes various external patent documentation files as part of its core workflow, which introduces a surface for indirect prompt injection. If an input file (e.g., CLAIMS.md) contains malicious instructions, the agent might attempt to follow them during the compilation process.
  • Ingestion points: The skill reads untrusted data from patent/CLAIMS.md, patent/specification/, patent/figures/, and patent/INVENTION_DISCLOSURE.md.
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent to ignore natural language instructions embedded within the source files.
  • Capability inventory: The skill has access to Bash(*), Read, Write, Edit, Grep, and Glob tools to process files and generate jurisdictional documents.
  • Sanitization: Absent. There is no evidence of validation or filtering for the content extracted from the input files before it is processed or written to new files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 03:14 AM
Security Audit — agent-trust-hub — jurisdiction-format