iterate
Warn
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill includes instructions that override standard agent behavior by mandating that the agent "NEVER ask 'should I continue?'" and instead proceed with autonomous actions until a stop condition is met.
- [COMMAND_EXECUTION]: The skill's logic is based on the autonomous and repeated execution of arbitrary shell commands defined in the 'Verify' and 'Guard' configuration settings.
- [PRIVILEGE_ESCALATION]: The instructions guide the agent to advise users on how to modify their security settings (.claude/settings.local.json) to provide wildcard permissions for the Bash tool, effectively bypassing the human-in-the-loop security prompts for command execution.
- [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface by executing unvalidated strings from the configuration as shell commands. 1. Ingestion points: 'Verify' and 'Guard' shell command strings in SKILL.md. 2. Boundary markers: None present. 3. Capability inventory: Bash, Write, Edit, and Agent tools. 4. Sanitization: None performed on user-provided command strings.
Audit Metadata