学习DNA档案
AdvisoryAudited by Static analysis on May 8, 2026.
Overview
No suspicious patterns detected.
Findings (0)
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.
If enabled, the assistant may remember learning goals, weak subjects, habits, mistakes, progress, and milestones across future sessions.
The skill explicitly creates cross-session persistent learning records. This is purpose-aligned and disclosed, but it stores sensitive student learning-profile data over time.
它深度依赖于 OpenClaw 平台底层的「本地化持久记忆（Localized Persistent Memory）机制」，将学习轨迹结构化沉淀在本地档案中
Enable it only if the student or guardian is comfortable with a persistent learning profile, and use the provided view, correction, deletion, and pause commands when needed.
Learning-profile details may be reused by other learning-related skills, such as reminders, review reports, mistake tracking, or study plans.
The skill is designed to share parts of the learning profile with named companion skills. The artifact limits this to listed skills and minimum fields, but cross-skill data flow is still privacy-relevant.
学习DNA档案 ├──→ 错题本（错误模式、弱项状态） ├──→ IM智能提醒（活跃时段、优先级、复习状态） ... 只共享完成当前任务所需的最小字段，不传整份档案
Review which skills can receive profile fields, and use the documented “不要共享” or pause controls if you do not want cross-skill sharing.
Certain student statements may cause the assistant to update the profile or involve another learning skill, rather than only answering in the current chat.
Reference guidance describes automatic memory updates and triggering related skills from user statements. The main SKILL.md adds consent, pause, and sharing boundaries, so this is a notice rather than a concern.
当学生说以下内容时，自动触发对应的DNA更新 ... "我这周要考XX" | 设置临时高优先级，触发IM提醒SKILL ... "这道题我做了三次都错" | 识别为固定错误模式，推送错题本SKILL
Make sure the user understands that these triggers operate only after profile consent, and honor “这次不要记忆” or “这次不要共享” requests.