page
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
score.pywithin the skill's own directory to validate landing page drafts. This execution is restricted by the skill's configuration to this specific script. - [INDIRECT_PROMPT_INJECTION]: The skill reads from
brand-config.jsonat the project root to populate its drafts. Ingestion points: Reads frombrand-config.jsonin the root directory. Boundary markers: No specific delimiters or safety instructions are used for the ingested data. Capability inventory: The skill is capable of writing to thegtm/directory and executing its own local validation script. Sanitization: Input data is mapped to specific JSON fields without explicit sanitization.
Audit Metadata