design-system
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads from an external reference file
../../references/DESIGN_SYSTEM.md. This represents an ingestion point where untrusted data could influence the agent's behavior if the file is attacker-controlled. - Ingestion points:
../../references/DESIGN_SYSTEM.md(Step 1) - Boundary markers: None identified in the instructions to separate external instructions from core logic.
- Capability inventory: The skill has the ability to write files (
06_release/artwork/design_system.json) and execute shell commands (scripts/design_system.py). - Sanitization: No evidence of sanitization or content validation is present in the instructions.
- [COMMAND_EXECUTION]: The skill instructs the agent to run a local script
scripts/design_system.pywith a generated JSON file as an argument. While this is a local script, it constitutes command execution based on processed data. - Evidence:
Run scripts/design_system.py 06_release/artwork/design_system.json(Step 6)
Audit Metadata