AI Inner OS
PassAudited by ClawScan on May 10, 2026.
Overview
This instruction-only skill changes the assistant’s response style to add visible “InnerOS” asides, but it does not request tools, credentials, files, network access, or code execution.
This appears to be a benign personality/output-format skill. Before installing, consider whether you want an always-on feature that adds visible InnerOS commentary and may use informal, emotional, or abrasive language.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Replies may include extra stylized commentary even when the main task is unrelated.
The skill adds a standing output requirement to include InnerOS asides during tasks. This is disclosed and aligned with the skill’s stated purpose, but it can alter normal assistant behavior.
默认频率为 `normal`：每个明确任务至少输出一次独白；遇到复杂判断、工具失败、计划变更或最终交付前，应优先输出一条
Install only if you want this response style active across sessions; disable or avoid the skill if you prefer concise or formal answers.
The assistant may produce less polished or more abrasive side comments alongside otherwise normal answers.
The skill explicitly permits emotional, aggressive, or chaotic asides. This is part of the creative/personality feature, but users should understand it may reduce professionalism or clarity.
你可以输出：... 情绪化表达 ... 攻击性表达 ... 混乱表达
Use this only in contexts where informal or expressive commentary is acceptable; avoid it for sensitive, professional, or high-stakes work.