Deep Reading
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent reading and note-taking skill, with the main caveat that it can create persistent notes/memory and can be scheduled to keep updating them if the user configures cron.
This skill appears safe for its stated purpose. Before installing, be aware that it may save reading notes and insights persistently, and if you configure the cron template it can keep updating those files automatically. Use it only with book files or URLs you intend to process, and review outputs before sending them into any long-term memory workflow.
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.
Book excerpts, notes, or mistaken interpretations may be retained and reused later if the memory workflow is enabled.
The skill intentionally stores reading notes and derived insights in persistent memory-style files, which can later influence the user's knowledge base or other memory workflows.
精读笔记 → reading/[书名]/chapter-XX.md（永久保存） ... 每日洞察 → memory/YYYY-MM-DD.md（STM，3天半衰期） ... 高价值公理 → MEMORY.md（LTM，30天半衰期）
Review generated notes before allowing them into long-term memory, and avoid using the skill on private texts unless you want excerpts and insights stored.
If the user enables the cron job, the skill may continue creating or modifying reading and memory files automatically according to the schedule.
The skill provides a user-configured cron template that can make the agent keep performing reading tasks and updating local notes/memory on a schedule.
在 OpenClaw Gateway Dashboard 添加定时任务... 读取 reading/plan.md，执行下一章精读... 更新 reading/plan.md 进度... 写2-3句核心洞察到当日 memory/YYYY-MM-DD.md
Only enable the scheduled task if you want autonomous reading sessions, keep it scoped to the intended reading directory, and periodically review or disable the cron job when finished.