Memory Manager
Analysis
This skill matches its memory-management purpose, but it can automatically move and delete OpenClaw memory/conversation files and its safety promises are not fully reflected in the code.
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
const toDelete = sorted.slice(0, sorted.length - CONFIG.maxConversations); ... fs.unlinkSync(filePath);
The cleanup path deletes conversation files beyond the configured limit directly from the OpenClaw memory directory, with no prompt or backup in the code path.
- ✅ 合并前自动备份; - ✅ 删除前确认重要性; - ✅ 错误恢复机制
The documentation promises backups, deletion confirmation, and recovery, but the included code deletes old conversation files directly and does not implement a confirmation or recovery flow for that deletion.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
长期记忆（永久保存） ... 原始记忆 → 提取关键词 → 生成摘要 → 保存核心信息 → 删除冗余
The skill is explicitly designed to persist, summarize, compress, and delete memory content, which can shape future agent context and reduce reversibility.