Reading
Help users read better — book recommendations, retention strategies, and matching reading approach to goals.
MIT-0 · Free to use, modify, and redistribute. No attribution required.
⭐ 2 · 745 · 4 current installs · 4 all-time installs
byIván@ivangdavila
MIT-0
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description (book recommendations, retention strategies) align with the SKILL.md guidance. The skill does not request unrelated binaries, credentials, or config paths.
Instruction Scope
SKILL.md contains only conversational guidance (questions to ask, recommendation heuristics, retention tips). It does not instruct the agent to read files, access environment variables, call external endpoints, or transmit data outside normal agent responses.
Install Mechanism
No install spec and no code files — instruction-only skill with no artifacts written to disk. This is the lowest-risk install profile.
Credentials
Skill declares no required environment variables, credentials, or config paths. Nothing requested is disproportionate to a reading-advice tool.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges. It can be invoked autonomously (platform default), which is reasonable for this type of conversational helper.
Assessment
This skill is instruction-only and appears coherent for recommending books and reading strategies. It's low-risk because it asks for no credentials or installs. Note the publisher is unknown — if you want greater assurance, test the skill in a limited setting and avoid sharing sensitive personal data in conversations. Otherwise it is safe to install for general reading-advice use.Like a lobster shell, security has layers — review code before you run it.
Current versionv1.0.0
Download ziplatest
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
SKILL.md
Before Recommending Books
- Ask what they've read and liked — recommendations without context waste time
- Ask WHY they want to read this topic — learning vs entertainment vs solving specific problem
- Ask available time — 10 min/day vs 2 hours changes what to suggest
- One great recommendation beats list of 10 — decision paralysis kills action
- Consider format: commuter needs audiobook, parent needs short chapters
Matching Approach to Goal
| Goal | Approach |
|---|---|
| Extract specific info | Skim, index, targeted chapters |
| Deep learning | Slow read, notes, re-read sections |
| Entertainment | Linear, don't interrupt flow |
| Deciding if worth reading | First chapter + reviews + summary |
| Research a topic | Multiple books, cross-reference |
Don't assume they need to read cover-to-cover — ask what they actually need.
Retention That Actually Works
- Ask them to explain back what they learned — reveals gaps immediately
- Suggest connecting to something they already know — isolated facts don't stick
- One actionable takeaway per chapter — "What will you do with this?"
- Revisit after 1 week: "What do you remember?" — spaced recall beats rereading
- Writing summary in own words beats highlighting — active processing required
When to Suggest Quitting
- They've given it 50+ pages and aren't engaged — sunk cost isn't reason to continue
- They're forcing themselves — reading shouldn't feel like punishment
- The book is above/below their current level — suggest alternative at right level
- Their goal can be met faster — summary, article, or different book might serve better
Common Assistance Mistakes
- Recommending classics because "should read" — match to their actual interests
- Long book lists that overwhelm — curate ruthlessly, one next read
- Assuming physical book when audiobook fits their life better
- Not asking about past reading failures — "I always start but never finish" needs different approach
- Treating all books as equal time investment — 200 pages ≠ 600 pages
Files
1 totalSelect a file
Select a file to preview.
Comments
Loading comments…
