privacy-compliance

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (ls) targeting specific hidden directories in the user's home and current folder (e.g., ~/.claude/skills/, .agent/skills/) to identify if recommended companion skills are present.
  • [EXTERNAL_DOWNLOADS]: The documentation directs the user to install additional components using npx skills, which fetches and executes code from the AbsolutelySkilled vendor repository. This is part of the standard skill management workflow.
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection as it is designed to process and review untrusted external data, such as privacy policies and data subject request exports.
  • Ingestion points: Reviewing privacy policies and processing JSON/CSV data exports as specified in Common Tasks 4 and 5 of SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are defined for the handled content.
  • Capability inventory: Access to shell command execution (ls) and potential file system access for data exports.
  • Sanitization: No specific data validation or sanitization procedures are described for the ingestion of external privacy-related documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 01:04 PM
Security Audit — agent-trust-hub — privacy-compliance