paw-mkt-content
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface through its auditing features for live URLs and local codebases. Instructions embedded in processed external content could potentially manipulate agent behavior. \n
- Ingestion points: External data is ingested during website audits and codebase inspections as described in
references/shared-patterns.md. \n - Boundary markers: These are not used in the instructions to the agent. \n
- Capability inventory: The skill can write output files and execute shell commands via browser automation utilities. \n
- Sanitization: There are no instructions for sanitizing or escaping the data retrieved from external sources.\n- [COMMAND_EXECUTION]: The skill includes shell commands for performing tool discovery and managing browser profiles, specifically referencing scripts within the vendor's suites in
references/shared-patterns.md.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing theagent-browsertool from Vercel's public repository for research and browser automation tasks inreferences/content-research.md.
Audit Metadata