paw-mkt-social

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the agent-browser utility from Vercel Labs' official GitHub repository to facilitate live competitor research and authenticated session management.
  • [COMMAND_EXECUTION]: Utilizes local shell scripts (tool-discovery.sh, chrome-profiles.sh) to verify the execution environment and identify available browser profiles for automated research tasks.
  • [DATA_EXFILTRATION]: The skill processes user-supplied brand identity and product context files from the local project workspace. While this creates a potential surface for indirect prompt injection, there are no patterns suggesting unauthorized data exfiltration to external endpoints.
  • Ingestion points: Files located in .pawbytes/marketing-suites/brands/{brand-slug}/ including brand-context.md, paw-mkt-product-context.md, and SOSTAC strategy documents.
  • Boundary markers: Not explicitly defined for external data ingestion.
  • Capability inventory: File system writing (deliverables), web browsing via agent-browser, and shell script execution for environment discovery.
  • Sanitization: Standard instructional prompts without explicit sanitization logic for ingested text.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 10:37 AM
Security Audit — agent-trust-hub — paw-mkt-social