moda-social-ads
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [SAFE]: The skill instructions and documentation are focused on legitimate design automation and do not contain malicious code, obfuscation, or persistence mechanisms. All tools belong to the author's verified namespace.- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it is designed to ingest and process data from user inputs, uploaded files, and external websites to inform design and generation tasks.
- Ingestion points: Campaign offer details, content from
moda file show, and data frommoda web readin the guidance for unbranded design. - Boundary markers: The recipes utilize standard CLI argument quoting but do not provide explicit instructions for the agent to ignore embedded commands in source data.
- Capability inventory: The skill has permission to execute shell commands (
moda:*), perform network requests, and mutate the design canvas. - Sanitization: Relies on the platform's host-level LLM guardrails and tool isolation.- [DYNAMIC_EXECUTION]: The documentation describes the use of
moda canvas editto run a restricted, synchronous JavaScript sandbox for canvas automation. The platform mitigates risk by enforcing a 100ms execution budget, a 16KB code size limit, and a restricted API that prevents persistence or destructive operations.
Audit Metadata