ck-help
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local script located at
.claude/scripts/ck-help.pyusing user-supplied$ARGUMENTS. This is the primary function of the skill to provide context-aware help. - [PROMPT_INJECTION]: The skill uses strong instructional markers such as 'IMPORTANT MUST ATTENTION' and 'CRITICAL' to enforce specific agent behaviors. These instructions focus on workflow integrity, such as breaking work into tasks with
TaskCreate, searching the codebase for patterns, and citing evidence for claims, rather than attempting to bypass safety filters or ignore system instructions.
Audit Metadata