serpapi
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill uses
bash -cto wrapcurlcommands. While this is used to address a specific environment bug and facilitate piping tojq, it involves executing shell commands which is a powerful capability. - [EXTERNAL_DOWNLOADS] (LOW): The skill communicates with
serpapi.comto fetch search results. This is the core functionality and targets a legitimate service, though it involves fetching external data. - [PROMPT_INJECTION] (LOW): The skill exhibits a surface for Indirect Prompt Injection (Category 8) because it retrieves search results (titles, snippets, and page content) from the public web.
- Ingestion points: Search result snippets and organic results from various search engines via SerpApi.
- Boundary markers: Absent; the skill returns filtered JSON but does not wrap content in delimiters or provide safety warnings to the agent.
- Capability inventory: The skill has access to the network via
curland executes shell commands viabash -c. - Sanitization: Results are filtered using
jq, which ensures structural integrity but does not sanitize the text content for potential malicious instructions.
Audit Metadata