hunt-business-logic
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and process data from external, untrusted web assets including
robots.txt,sitemap.xml, and JavaScript bundles. This behavior establishes an attack surface where a malicious target site could serve content containing natural language instructions designed to influence or hijack the agent's logic. - Ingestion points: Network requests to target-controlled files such as
https://target.com/robots.txtandhttps://target.com/assets/app.js(found inSKILL.md). - Boundary markers: The skill does not provide instructions to the agent to use delimiters or to ignore potential instructions embedded within the data retrieved from external targets.
- Capability inventory: The skill utilizes network access (
curl) and shell-based data processing (grep) to interact with and analyze external targets. - Sanitization: There is no evidence of sanitization or strict schema validation for the external content before it is processed by the agent.
Audit Metadata