book-review
Skill1 starsChanged 5 months agoMITView on GitHub
Document a reader's experience with a non-fiction book through active recall and guided reflection. Triggers when user wants a book review/summary/notes AND provides their highlights, annotations, or reading data (JSON, HTML, CSV, export, copy-paste). Interviews the user with four reflection questions before evaluating their recall against their own highlights, checking factual accuracy, and flagging missed themes. Produces a one-page summary in the user's voice. Do not trigger for fiction, or requests without the user's own reading data.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- CSPulse/customer-success-skillstheir own version, first seen Nov 8, 2026
- holutahyour/claude-skillstheir own version, first seen Oct 20, 2026
- kyle-olmstead/cs-claude-enginetheir own version, first seen Oct 17, 2026
- tachyon-zcash/ragutheir own version, first seen Oct 8, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- tachyon-zcash/ragu★ 61same name
- CSPulse/customer-success-skills★ 2same name
- kyle-olmstead/cs-claude-engine★ 1same name
- holutahyour/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 13 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
13
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
Do not trigger for:
- Distinctive: line 52
Ask all four questions at once:
- Distinctive: line 63
Wait for the user's full response before proceeding.
- Distinctive: line 67
After receiving answers, assess against three sources:
- Distinctive: line 72
Check if stated "core argument" aligns with what they actually marked
- Distinctive: line 75
Check for inaccuracies in their recall
- Distinctive: line 79
Search reliable sources (publisher descriptions, reputable reviews, author interviews) for major themes
- Distinctive: line 80
If they missed a commonly-discussed theme, flag it: "Critics often highlight X as a central theme — your highlights touch on this in …
- Distinctive: line 90
Power stays with the user: After flagging gaps, ask: "Any of these worth adding to your summary, or are you happy with your current …
- Distinctive: line 94
Create a one-page summary combining:
- Distinctive: line 140
Recall before assistance: Always interview before evaluating.
- Distinctive: line 146
No promotional language: Write clean, analytical prose.
- Distinctive: line 162
If format is unclear, ask the user to clarify structure.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Book Review Skill
Introduction
Purpose
This skill helps readers document their experience with a book — not produce a generic summary.
A book review should capture what you noticed, what your mind caught onto, what resonated with your context. The interactive interview exists because recall is the point: articulating what you remember cements learning. AI fills gaps, not replaces thinking.
The goal is threefold:
- Prompt active recall — force articulation before showing what was highlighted
- Surface blind spots — compare stated understanding against actual annotations
- Document the reader's experience — produce a summary that reflects their perspective, not a Wikipedia entry
This skill should reduce over-reliance on AI by making the user do the work first.
When to Apply
Trigger this skill when:
- User mentions wanting a book review, book summary, or book notes
- User provides reading data: highlights, annotations, bookmarks, or notes
- Format can be anything: JSON, HTML, CSV, export, plain copy-paste
Do not trigger for:
- Fiction books (this workflow assumes non-fiction with arguments/themes)
- Requests for a summary without the user's own reading data
- Academic citation or bibliography tasks
What This Skill Governs
Workflow
1. Intake
Parse the user's reading data. Extract:
- Highlights/quotes
- Bookmarks (chapter locations)
- User annotations/notes
Confirm book title and author. If unclear, ask.
2. Interview
Ask all four questions at once:
Before I evaluate your notes, I'd like to hear your recall. Answer as briefly or thoroughly as you like:
1. What was the core argument of the book? What were the key supporting points?
2. What did you find insightful?
3. What were you sceptical about?
4. How can you apply what you learned?Wait for the user's full response before proceeding.
3. Evaluate
After receiving answers, assess against three sources:
A. Their highlights
- Identify themes they highlighted but didn't mention
- Note bookmarked chapters they didn't discuss
- Check if stated "core argument" aligns with what they actually marked
B. Factual accuracy
- Check for inaccuracies in their recall
- Correct gently with evidence from their own highlights or reliable sources
C. External themes
- Search reliable sources (publisher descriptions, reputable reviews, author interviews) for major themes
- If they missed a commonly-discussed theme, flag it: "Critics often highlight X as a central theme — your highlights touch on this in Chapter Y, but you didn't mention it. Worth revisiting?"
Tone: Gentle nudges. The user decides what matters to them.
Examples:
- "You highlighted several passages about X but didn't mention it — anything there?"
- "Your heaviest annotations were in the section on Y. Deliberately omitted, or slipped your mind?"
- "One factual note: the book states Z happened in 1974, not 1971."
- "Reviewers frequently cite the author's argument about W as central. You bookmarked that chapter but didn't mention it — intentional?"
Power stays with the user. After flagging gaps, ask: "Any of these worth adding to your summary, or are you happy with your current framing?"
4. Output
Create a one-page summary combining:
- Core thesis — from user's response, supplemented by highlights
- Key insights — what they found valuable + overlooked highlights
- Critiques — their stated skepticism
- Applications — how they'll use it
- Gaps noted — brief mention of themes they chose to omit (optional, user decides)
5. Export
Offer:
- Markdown file
- Word document (.docx) for Google Drive — follow
/mnt/skills/public/docx/SKILL.md
Examples
Correct Usage
User: "I just finished Chokepoints by Edward Fishman. Here are my highlights: [JSON link]. Can you help me create a book review?"
Claude: Parses JSON. Confirms title/author. Asks the four questions. Waits.
User: Responds to all four.
Claude: Evaluates recall against highlights. Notes: "You highlighted extensively on the petrodollar mechanics but didn't mention it — worth including?" Checks reliable sources, flags one theme the user missed. Asks if they want to incorporate any gaps. Produces one-page summary in user's writing style. Offers export.
Incorrect Usage
| Scenario | Why it's wrong |
| Producing a summary immediately without asking the four questions | Skips the recall step; defeats the purpose |
| Writing in generic "book report" style | Should match user's voice and style |
| Overwhelming user with every gap found | User decides what matters; flag, don't lecture |
| Using promotional language ("must-read," "brilliant") | Analytical, not promotional |
| Searching extensively to pad the summary | Search only to verify facts or flag missed themes |
| Triggering for fiction or books without user's reading data | Skill assumes non-fiction + user annotations |
Principles
- User's voice first. Match their writing style. The summary reflects their experience, not a generic overview.
- Recall before assistance. Always interview before evaluating. The act of articulation is the learning.
- Highlights reveal attention. What someone marks shows what resonated. Use this as ground truth for evaluating recall.
- Gaps are suggestions, not corrections. Flag what they missed. They decide what to keep.
- No promotional language. Write clean, analytical prose. No "definitive," "essential," "groundbreaking."
- Brevity over completeness. One page means tradeoffs. Prioritize what the user cared about.
Format Handling
| Format | Parsing approach |
| JSON | Parse highlights, bookmarks, notes arrays |
| HTML | Extract highlighted text, annotations |
| CSV | Map columns to quotes/notes/chapters |
| Export | Parse clippings format |
| Plain text | Treat as raw highlights; ask for clarification if ambiguous |
If format is unclear, ask the user to clarify structure.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

