seo-drift
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill body contains instructions that direct the agent to read a specific file path (
../../seo/seo-drift/SKILL.md) and "follow it as the active workflow." This pattern attempts to redirect the agent's control flow and override the current instructions with content from a different location. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes external website data based on user-provided URLs, creating a vulnerability surface where malicious instructions could be embedded in the processed data.
- Ingestion points: External site metadata, titles, meta descriptions, and schema presence fetched via the user-supplied URL (SKILL.md).
- Boundary markers: The skill lacks explicit delimiters or instructions for the agent to ignore potentially malicious content found within the analyzed web data.
- Capability inventory: The skill captures, snapshots, and diffs site state, which involves parsing untrusted strings from external web pages.
- Sanitization: No mention of content sanitization or filtering of external inputs is present in the skill instructions.
Audit Metadata