marketing-seo-specialist
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates tools such as WebFetch and WebSearch to ingest and process data from external websites. This architectural pattern introduces a vulnerability surface for indirect prompt injection, where malicious instructions hidden in third-party web content could attempt to influence the agent's internal logic or behavior.\n
- Ingestion points: The tools WebFetch and WebSearch are defined as primary capabilities in the frontmatter of SKILL.md.\n
- Boundary markers: Absent. The skill instructions do not provide specific delimiters or 'ignore' directives to help the agent distinguish between its core SEO mission and potentially malicious instructions embedded in the data it retrieves.\n
- Capability inventory: The agent has a broad capability set including WebFetch, WebSearch, Read, Write, and Edit, which could be leveraged if an injection is successful.\n
- Sanitization: There is no evidence of content filtering, validation, or sanitization logic to be applied to the output of the web-fetching tools.\n- [NO_CODE]: The skill consists entirely of markdown-based identity instructions and templates. No Python scripts, JavaScript files, or shell commands are included, which eliminates the risk of direct remote code execution or system-level compromise from the skill's own assets.
Audit Metadata