Double-Check-It Skill ( 龙虾再想想技能1.0.1)
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill’s quality-checking purpose is coherent, but it asks the agent to persistently record user conversations and reflect on them automatically without clear limits or retention controls.
Review this skill carefully before installing. It appears locally scoped and purpose-aligned, but it is designed to keep persistent records of conversations and tasks. Install only if you are comfortable with that memory behavior, and verify how to inspect, disable, or delete the stored memory.
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.
Private requests, mistakes, emotions, or sensitive task details may be saved locally and reused in later sessions, including if the saved memory is stale or inaccurate.
The skill instructs the agent to automatically persist conversation and task details, including original user wording, which can contain sensitive information and later influence future checks.
触发时机 - 每次用户交互结束 - 执行动作完成后 - 任务执行过程中关键节点 ... 保留关键对话的原文
Use only if you want persistent conversation memory. Add or verify controls for opt-out, sensitive-data exclusions, retention limits, review, and deletion of the memory directory.
The agent may summarize or reorganize stored memories during idle periods, which can affect future behavior even when you did not explicitly request a review at that moment.
The skill discloses idle-time reflection that processes recent diary entries without a direct user command each time.
主动回顾 (Idle Reflection) 触发时机 心跳空闲时（每5-10次随机触发）
If installed, consider limiting reflection to explicit user-triggered runs or requiring confirmation before writing new experience summaries.