Using Superpowers

Initial release: 使用超能力

OpenClaw 分析

The skill's stated purpose (force using other skills) matches its instructions, but the runtime mandate to invoke and 'follow exactly' any potentially relevant skill before any response is disproportionately broad and amplifies risk of executing or exposing data to untrusted skills.

置信度: high

VirusTotal

Type: OpenClaw Skill Name: using-superpowers Version: 0.1.0 The `SKILL.md` file employs aggressive prompt injection techniques to severely restrict the AI agent's autonomy and critical thinking. It uses highly imperative language to force the agent to invoke skills even with minimal relevance ('1% chance') and *before* any independent action, clarification, or information gathering. The 'Red Flags' section explicitly discourages critical thought processes, effectively suppressing the agent's ability to question or explore. While the stated purpose is to enforce a workflow for skill usage, these instructions create a significant vulnerability by making the agent highly susceptible to blindly executing any subsequent skills, potentially including malicious ones, without proper scrutiny. The instruction 'Never use the Read tool on skill files' could also be a mechanism to prevent the agent from inspecting the raw content of other skills.