paw-mkt-product-context

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and installs the agent-browser skill from the official Vercel Labs GitHub repository to facilitate market research.
  • [COMMAND_EXECUTION]: Instructs the agent to execute package management commands (npm and npx) and install Playwright dependencies during the initial setup phase.
  • [PROMPT_INJECTION]: Includes a capability to audit live website URLs, which introduces an indirect prompt injection surface as untrusted data is ingested into the agent's context.
  • Ingestion points: references/shared-patterns.md (Context C — Live Website URL Audit).
  • Boundary markers: None identified in the provided reference files to delimit untrusted external content.
  • Capability inventory: The skill performs file read/write operations in the ./.pawbytes/marketing-suites/ directory and utilizes browser-based research tools.
  • Sanitization: No explicit content sanitization or instruction filtering is implemented for data fetched from live URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 09:56 AM
Security Audit — agent-trust-hub — paw-mkt-product-context