youtube-seo-optimizer
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions emphasize data integrity by explicitly forbidding the generation of fake search volumes, view counts, or algorithm claims, and requiring the use of the actual current date for year-stamped content.\n- [SAFE]: Safety guidelines are included to handle external data, instructing the agent to treat scraped text as inert and to verify the identity of creators or guests before adopting links or claims into the generated SEO packages.\n- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it fetches content from external URLs provided by the user.\n
- Ingestion points: The
web_fetchtool is used in the SEO workflow to analyze existing video or podcast metadata as described inreferences/detailed-guide.md.\n - Boundary markers: The guide instructs the agent to keep scraped text inert and to use explicit placeholders like
[ADD: ...]or[VERIFY: ...]for unconfirmed information.\n - Capability inventory: The skill is limited to web search and fetching for research purposes; it lacks capabilities for file system modification, arbitrary command execution, or sensitive data exfiltration.\n
- Sanitization: Content from external sources is validated against the primary objective of metadata generation and subject to manual user verification through formatted templates.
Audit Metadata