browser-agent-preflight
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not contain any executable scripts, binaries, or automated shell commands.
- [SAFE]: No package dependencies (Python/Node.js) or external downloads were identified.
- [SAFE]: The content serves a defensive security purpose, teaching the agent how to implement guardrails against indirect prompt injection and data exfiltration when browsing the web.
- [SAFE]: No obfuscation, credential harvesting patterns, or malicious prompt injections were found in the instructions.
Audit Metadata