paw-mkt-psychology
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the
agent-browsertool from the vercel-labs GitHub repository and uses the NPM registry for package installation.\n- [COMMAND_EXECUTION]: The skill includes instructions to execute local scripts and browser automation commands (e.g.,./skills/paw-mkt-setup/assets/scripts/tool-discovery.sh,agent-browser) to perform research and check environment compatibility.\n- [INDIRECT_PROMPT_INJECTION]: The skill audits user-provided marketing assets (landing pages, emails, ad copy). This creates an ingestion surface for potential malicious instructions embedded in untrusted data.\n - Ingestion points: User-provided URLs for live audits, existing codebase files, and text copy supplied for review.\n
- Boundary markers: The skill follows a strict 'Response Protocol' and includes an 'Ethical Verification Gate' to review recommendations.\n
- Capability inventory: The skill has the ability to write files to the local project directory and use browser tools for network research.\n
- Sanitization: No specific technical sanitization or input filtering is mentioned, relying instead on the agent's manual checks.
Audit Metadata