caption-writer-codex
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions for text generation and content formatting.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from several local markdown files (e.g., brand-style.md, content-calendar.md). While these external files are technically untrusted inputs, the skill does not possess any dangerous capabilities—such as shell execution, network access, or file system writing outside of its designated output folder—that could be exploited via injection. This constitutes a standard retrieval-augmented generation workflow.
Audit Metadata