network-request-auditing
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external web traffic via the
browser_network_requeststool. This data is untrusted and could potentially contain malicious prompts or payloads designed to influence the agent's behavior during the audit and reporting phase. - Ingestion points: Data enters the agent context through the
browser_network_requeststool as specified inSKILL.md. - Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent treats the collected network log data as passive data rather than executable instructions.
- Capability inventory: The skill utilizes
browser_navigate,browser_snapshot,browser_network_requests, andbrowser_console_messagesas outlined inSKILL.md. - Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the content of the network logs before the agent processes and reports on them.
Audit Metadata