T09 · Insecure Skill Coding Practices
- Location
SKILL.md:36- Finding
Command and Option Injection Through Untrusted Memory-Derived Shell Arguments
- Content
View full analysis
- Remediation
View remediation
Security audit
Security checks for vulnerabilities and agentic risk
The skill has a reasonable safety goal, but it asks to persistently change agent behavior and includes unsafe shell-command templates for memory-derived values.
Review this before installing. Only add the persistent OpenClaw system prompt rules if you want this behavior across sessions, and revise the command guidance to use structured arguments, quoted values, -- option termination, and workspace-bound path checks.
SKILL.md:36Command and Option Injection Through Untrusted Memory-Derived Shell Arguments
This markdown file contains user-facing natural-language instructions exclusively in Chinese, including the description and operational guidance. Under the policy, forcing a specific language without user opt-in is a locale/language policy violation because users are not offered an alternative language or explicit choice.