gws-script

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it can ingest untrusted data from external sources.
  • Ingestion points: The getContent method in SKILL.md allows the agent to read the source code and metadata of script projects.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the skill definition.
  • Capability inventory: The skill provides capabilities to write code (updateContent) and execute scripts (run) based on interpreted instructions.
  • Sanitization: There is no evidence of sanitization or validation of the ingested script content before it is processed or used to inform subsequent actions.
  • [COMMAND_EXECUTION]: The skill requires and executes a local binary named gws. This appears to be the primary tool provided by the vendor for interacting with Google Workspace APIs and is used for management tasks like listing processes and updating project content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — gws-script