Kaiqiao

PassAudited by ClawScan on May 10, 2026.

Overview

This skill is a behavior-coaching prompt that is mostly transparent, but it asks the agent to reduce confirmations, remember preferences, and set periodic follow-up reminders.

Install only if you want your agent to be more proactive, ask fewer confirmations in authorized situations, and remember communication preferences. Keep explicit approval requirements for irreversible or high-impact actions, and periodically review any USER.md, MEMORY.md, or recurring reminders the skill creates.

Findings (3)

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.

What this means

The agent may act faster and ask fewer confirmation questions, which can be helpful for small reversible tasks but should not replace explicit approval for high-impact actions.

Why it was flagged

The skill deliberately reduces repeated confirmations after perceived authorization. That is consistent with the behavior-calibration purpose, but it could affect how cautiously the agent proceeds with future actions.

Skill content
授权信号 — 对方说"你觉得有用就做"/"你看着办"/"行,搞吧" ... 输出格式:无(直接执行)
Recommendation

Use this skill with clear boundaries: tell the agent which actions still require confirmation, such as payments, deletions, publishing, account changes, or irreversible operations.

What this means

Personal communication preferences and work habits may be saved and reused later, including if they become outdated or are recorded incorrectly.

Why it was flagged

The skill instructs the agent to store learned user preferences and lessons in persistent memory files. This is aligned with personalization, but persistent memory can affect future agent behavior.

Skill content
对方的偏好、习惯、口头禅含义 | USER.md ... 任务中踩的坑、纠正的教训 | MEMORY.md
Recommendation

Review USER.md and MEMORY.md periodically, delete outdated entries, and avoid saving sensitive personal or business details unless necessary.

What this means

The agent may periodically surface reminders or preference checks after installation.

Why it was flagged

The skill asks for a recurring reminder that causes the agent to proactively revisit stored preferences. This is disclosed and purpose-aligned, but it introduces ongoing behavior beyond a single interaction.

Skill content
每隔1-2周,Agent 会主动确认共识有没有过时 ... 必须设一个定期提醒,不能靠自觉
Recommendation

Only enable the recurring reminder if you want ongoing follow-ups, and make sure you know where to disable or delete the reminder.