glossary-resolver

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a glossary resolution feature using standard Node.js patterns and secure file IO wrappers provided by the @agent/core library.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted input files and glossary definitions without sanitization.
  • Ingestion points: Reads document content and glossary data from paths provided via command-line arguments in src/lib.ts and scripts/resolve.ts.
  • Boundary markers: No markers or warnings are used in the output to separate original content from resolved glossary definitions.
  • Capability inventory: Access is limited to reading and writing local files; no network or shell execution capabilities were detected.
  • Sanitization: Input content and glossary definitions are used directly in string replacement without validation or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 06:28 PM