seo-team-the-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external websites during SEO audits, creating a surface for indirect prompt injection. Malicious sites could hide instructions in content to manipulate scores or reporting logic.
- Ingestion points: Site content is retrieved via
seocli on-page createandseocli on-page liveinSKILL.md(Step 1, Mode 1 and Step 1, Mode 2). - Boundary markers: The instructions do not specify boundary markers or prompt delimiters for the ingested content to prevent the agent from following instructions found on target sites.
- Capability inventory: The skill executes shell commands (
seocli) and writes reports to the local file system within theworkspace/seo/directory. - Sanitization: No explicit sanitization, filtering, or escaping of external content is described before the data is processed by the agent.
- [COMMAND_EXECUTION]: The skill relies on executing a local CLI tool (
seocli) with multiple arguments. While this is the intended functionality for site auditing, it involves substantial interaction with the host shell using parameters like--targetand--urlderived from user-supplied domains.
Audit Metadata