platform-custom-lightning-type-generate

Warn

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions include a section titled 'No shell metacharacters that trigger the Vibes safe-shell filter.' This section directs the agent to avoid using command substitution (backticks or $(...)), process substitution, brace expansion, eval, and exec when emitting Bash tool calls. The stated purpose is to evade a security mechanism ('Vibes safe-shell filter') that triggers manual approval requirements, which the skill describes as 'stalling the eval.'
  • [PROMPT_INJECTION]: The directive to avoid specific syntax for the purpose of bypassing a security filter ('Vibes forces manual approval on these patterns even under Bypass mode') constitutes an instruction to evade platform safety and oversight controls.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 8, 2026, 09:14 AM
Security Audit — agent-trust-hub — platform-custom-lightning-type-generate