serp-keyword-research
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from various search engines (Google, Bing, Brave, DuckDuckGo, Yahoo) and Google Trends via the Crawlora API. This data is untrusted and could contain instructions designed to manipulate the AI agent.
- Ingestion points: Data enters through the
scripts/crawlora.shhelper which makes requests to the endpoints defined inreference/endpoints.md. - Boundary markers: The output is structured as JSON, but there are no specific prompt delimiters or instructions to the agent to treat this external content as untrusted.
- Capability inventory: The skill itself facilitates network requests to the vendor API; however, the agent using this skill might have other capabilities like file writing or code execution that could be targeted by an injection.
- Sanitization: The helper script uses
jqto process the JSON output, but does not perform semantic validation or sanitization of the search result text.
Audit Metadata