费曼学习法
PassAudited by VirusTotal on May 8, 2026.
Overview
Type: OpenClaw Skill Name: xiaozhi-feynman-test Version: 1.1.0 The skill bundle implements a pedagogical framework based on the Feynman technique, designed to help students verify their understanding through explanation and Socratic questioning. The instructions in SKILL.md and the dialogue patterns in references/feynman-dialogue-patterns.md are entirely focused on educational interaction, including age-appropriate logic and progress tracking via a 'learning DNA' profile. No indicators of data exfiltration, malicious execution, or harmful prompt injection were found.
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.
The agent may keep records of a student’s understanding level, weak points, and suggested follow-up actions across sessions.
The skill directs the agent to persist each test result into a learner profile, creating ongoing educational memory.
每次测试结束后，都要写回DNA中的理解深度档案，而不是只给口头反馈。
Use it only if you are comfortable with persistent learning records; check how the linked DNA/profile system stores, edits, and deletes those records.
A student’s weak areas and mastery status may be reused by other study features beyond the immediate Feynman-test conversation.
The skill describes sharing learning-status data with other skills or agents for reminders, weekly review, and coordination.
──→ IM智能提醒（发现盲区后的复测提醒）
──→ 每周学习复盘（本周理解深度数据）
──→ 五SKILL协调器（联动判断与月报汇总）Confirm which related skills can receive this data and whether reminders or reports are visible to parents, teachers, or other users.
This skill’s actual profile read/write behavior may depend on a separate component that has not been reviewed here.
The skill declares reliance on another skill for learner-profile behavior, but that dependency’s contents are not part of the provided review artifacts.
depends_on: xiaozhi-learning-dna
Review the xiaozhi-learning-dna skill separately before relying on the persistent profile features.