keyword-research
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
bashtool to runcurlandjqcommands to interact with SEO APIs. This is standard behavior for the tool's purpose. - [EXTERNAL_DOWNLOADS]: Retrieves data from well-known SEO service providers including SemRush, DataForSEO, and SerpAPI.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external search results and API responses.
- Ingestion points: Search result snippets and 'People Also Ask' questions are retrieved from
serpapi.com,api.semrush.com, andapi.dataforseo.comvia instructions inSKILL.md. - Boundary markers: No delimiters or isolation instructions are present to prevent the agent from following instructions embedded in the API data.
- Capability inventory: The skill utilizes the
bashtool, which could be exploited if malicious instructions are successfully injected. - Sanitization: No sanitization or content filtering is implemented for the external data.
Audit Metadata