paw-mkt-content
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to fetch and install the
agent-browserutility from thevercel-labsorganization on GitHub. - [PROMPT_INJECTION]: The skill supports a 'Live Website URL Audit' mode where it audits content from external web pages. This creates a surface for indirect prompt injection, as the agent is not explicitly instructed to ignore or isolate potential instructions embedded within the untrusted external data during its audit.
- [COMMAND_EXECUTION]: The skill utilizes shell-based commands such as
npmandnpxfor environment setup and tool installation, as well as theagent-browserCLI for executing web research tasks.
Audit Metadata