delete-render-profile

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes shell commands (jq, mv, cat, test) to manage a local JSON file in a designated application data directory.
  • [PROMPT_INJECTION]: The skill processes and displays contents from render-profiles.json, which presents an indirect prompt injection surface.
  • Ingestion points: Reads from render-profiles.json in SKILL.md.
  • Boundary markers: No specific delimiters or instructions provided to the agent to distinguish between data and potential commands within the file.
  • Capability inventory: Access to Bash (test, jq, mv, cat), Read, and Write tools across the script.
  • Sanitization: The skill uses the jq --arg flag to safely interpolate the profile name, preventing shell injection into the jq command logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:35 PM
Security Audit — agent-trust-hub — delete-render-profile