2025-04-13 | 💪 Empower 👥 People

📚 Books

🤖🤖📖 An AI Book Report Template

💻 This is the Obsidian templater template I’m currently using for AI book reports.
➕ I use the QuickAdd plugin to call the template.
📱 I use the Commander plugin to put a button on my mobile toolbar
🛠️ It’s not a perfect work of software craftsmanship; just a practical script that streamlined and automates.
✅ I think it’s working pretty well!
⚙️ Workflow
- ➕ Click the plus button on my mobile toolbar (bringing up the QuickAdd menu)
- 📖 Choose book from the menu
- 📝 Paste the name of a book (no colons due to filename restrictions) into the text box
- ⏳ Wait several long seconds (less than a minute, but there’s a definite time delay)
- 💰 Profit!
🤖 The template will
- 📄 create a new file named after the book (in kebab-case)
- ⚙️ set up some frontmatter (yaml at the top of the markdown file) and boilerplate text and links
- 🗣️ Call the Gemini API asking for a book report (see the prompt below)
  - 💰 using a newer model with a relatively low rate limit & daily cap
- ✨ Send the result back to Gemini to Emojify it
  - ⚡ using a good standard model with a very generous rate limit and daily cap
- 🧹 Clean up some whitespace
- ✏️ Insert the result into the new note
- 🔗 (QuickAdd adds a link to the new file wherever my cursor was)
✍️ After that, I may do some light editing, pick some nice emojis for the title, etc - but it’s largely ready to go
🔑 All you need is a (free) Gemini API key.

<%*  
const filename = tp.file.title  
const kebabCase = filename  
  .toLowerCase()  
  .replaceAll("'", '')  
  .replace(/[^a-z0-9]+/g, '-')  
  .replace(/^-+|-+$/g, '')  
%>---  
share: true  
aliases:  
  - "<% filename %>"  
title: "<% filename %>"  
URL: https://bagrounds.org/books/<%kebabCase%>  
Author:   
tags:  
---  
[[index|Home]] > [[/books/index|Books]]  
# <% filename %>  
<%*  
async function generateTextFromGemini(prompt, apiKey, model = "gemini-2.0-flash", search = false) {  
  const apiUrl = `https://generativelanguage.googleapis.com/v1beta/models/${model}:generateContent?key=${apiKey}`;  
  
  const body = {  
      "contents": [  
          {  
              "parts": [  
                  {"text": prompt }  
              ]  
          }  
      ],  
      "tools": [  
          {  
              "google_search": {}  
          }  
      ]  
  };  
  if (!search) delete body.tools  
  
  
    const response = await fetch(apiUrl, {  
      method: "POST",  
      headers: {  
        "Content-Type": "application/json",  
      },  
      body: JSON.stringify(body),  
    });  
  
    if (!response.ok) {  
      throw new Error(`HTTP error! status: ${response.status}`);  
    }  
  
    const data = await response.json();  
    const { parts } = data.candidates[0].content;  
    return (parts[1] || parts[0]).text;  
}  
  
const GEMINI_API_KEY = "https://ai.google.dev/gemini-api/docs/api-key"; // Replace with your actual API key.  
  
const prompt = `Write a markdown-formatted (start headings at level H2) book report, followed by a plethora of additional similar, contrasting, and creatively related book recommendations on ${filename}. Be thorough in content discussed but concise and economical with your language. Structure the report with section headings and bulleted lists to avoid long blocks of text.`;  
  
const emojifyPrompt = report => `Enhance the selected markdown content by inserting relevant emojis: "${report}".  
Each emoji should go before the word, phrase, or sentence it represents, followed by a space.  
Put at least 1 emoji at the beginning of each section heading, bullet point, or line of text.  
Insert emojis in long blocks of text.  
Only return the text with the emojis.  
Ensure the placement of emojis doesn't conflict with markdown syntax.`;  
  
const model = 'gemini-2.5-pro-exp-03-25'  
// const model = 'gemini-2.0-flash-thinking-exp-01-21'  
  
try {  
    const report = await generateTextFromGemini(prompt, GEMINI_API_KEY, model, true);   
    const emojified = await generateTextFromGemini(emojifyPrompt(report), GEMINI_API_KEY);  
  
    const clean = emojified  
        .replaceAll(/(?<!^ *)( +)/gm, ' ')  
        .replaceAll(/ +$/gm, '')  
        .replaceAll(/\n\n\n+/g, '\n\n')  
  
    tR += clean;  
} catch (error) {  
    tR += `Error: ${error.message}`;  
}  
%>  
  
## 💬 [[software/gemini|Gemini]] Prompt (<% model %>)  
> <% prompt %>.  
  
<%*  
await tp.file.rename(kebabCase)  
%>

🤪 Emojify Template

🤔 Choosing and 🖱️ inserting the 👌 perfect emojis can be 😫 tedious.
🥳 Luckily, ~~😈🤖Skynet~~ 😇🤖 Gemini is pretty 👍 good with emojis.
Similar to above, I use 👨‍✈️ Commander to put a 🖲️ button with a 🎨 custom icon on my 📱 mobile toolbar
👆 Select text. 🙂 Tap button. ✨ Emojify.

<%*  
async function generateTextFromGemini(prompt, apiKey, model = "gemini-2.0-flash") {  
  const apiUrl = `https://generativelanguage.googleapis.com/v1beta/models/${model}:generateContent?key=${apiKey}`;  
  
  const requestBody = {  
    contents: [{ parts: [{ text: prompt }] }],  
  };  
  
  try {  
    const response = await fetch(apiUrl, {  
      method: "POST",  
      headers: {  
        "Content-Type": "application/json",  
      },  
      body: JSON.stringify(requestBody),  
    });  
  
    if (!response.ok) {  
      throw new Error(`HTTP error! status: ${response.status} - ${JSON.stringify(await response.json(), null, 1)}`);  
    }  
  
    const data = await response.json();  
    return data.candidates[0].content.parts[0].text;  
  } catch (error) {  
    return `Error: ${error.message}`;  
  }  
}  
  
const GEMINI_API_KEY = "https://ai.google.dev/gemini-api/docs/api-key"; // Replace with your actual API key.  
const selectedText = tp.file.selection();  
  
if (selectedText) {  
  const prompt = `Enhance the selected text by inserting relevant emojis: "${selectedText}".  
  Each emojis should go before the word, phrase, or sentence it represents, followed by a space.  
  Put at least 1 emoji at the beginning of each section heading, bullet point, or line of text.  
  Insert lots of emojis in long blocks of text.  
  Only return the text with the emojis.`;  
  
  const emojifiedText = await generateTextFromGemini(prompt, GEMINI_API_KEY);  
  if (emojifiedText) {  
      tR = emojifiedText;  
  } else {  
    tR += "Failed to emojify text.";  
  }  
} else {  
  tR += "Please select some text to emojify.";  
}  
%>

🪞 Reflections

❓ How much value can we get from an AI generated 🤖 book report 📝?
- 🤔 Maybe 10% of the value of the full read in …
  - 📚 Strongmen: Mussolini to the Present is 9h 42m on audible. 🎧
  - ⏱️ The read time plugin claims my AI book report 📝 is a 7 minute read.
  - 🧮 (9 * 60 + 42 = 582) / 7 = 83
- ⏳ … 1/83rd the time?
- 🚀 That’s something like 8x leverage.
- 👍 Not bad.
- 🎁 And the Gemini API is free.
🌐 How much value can be created by freely sharing these AI book reports 📝 online?
- 📈 According to the Google Search Console, traffic 🖱️ (clicks when it comes up in search results) to my website 💻 has grown from 3 per week to 23 per week over the past 3 months.
- 📊 That’s a growth rate of (23 - 3) / 3 ~= 7 additional clicks per month.
- 🔮 Extrapolating with linear growth, I might have (23 + 36 x 7 =275) clicks per week 3 years from now.
- ➕ Integrating (summing) over time, that yields (Gemini pause for LaTex) $\sum_{i = 0}^{36} (23 + 7 * i)$ = 5513 (Thanks Wolfram Alpha!) clicks over the next 3 years.
  - ☝️🐛 See the bug in this calculation?
- 🤓 Assuming 1% of my visitors actually read a single AI generated 🤖 book report 📝 (and the rest read nothing), that’s 55 AI book reports read.
- 🤯 …do I get to multiply my earlier 8x leverage per book 📚 by 55 people 🧑‍🤝‍🧑, yielding 440x?
- ❓ What units does this leverage take on? Book 📖 knowledge dissemination rate?
- 🤔 Not sure…
- 🗳️ Maybe that’s 55 more slightly better informed voters by our next election?
- ❤️‍🩹 That probably won’t do much on its own, but it’s a contribution.
🤪 Emojify! 🪄
🧐 Reviews the above…
- 🐛 Bug report: I mixed weeks and months! 🗓️
- 🧮 I shouldn’t be summing to 36. 🤔
- 🗓️ It should be 3 years x 12 months per year x 4 weeks per month (or more accurately, 📅 52 weeks per year x 3 years) = 156
- 📈 So $\sum_{i = 0}^{156} (23 + 7 * i)$ = 89,333 ! 🤯
- 👓 With the 1% reading rate, we now get 893 more slightly better informed voters by next election. 🗳️
- 😞 Which, to be honest, is qualitatively about as disappointing as 55. 😔
- 📍 And they’re probably not in swing states. 🗺️
- 💻 Though maybe I could check readership by state in the Google Search Console. 😴 Not tonight though.
🎯 Okay, maybe my next goal should be to figure out how to inject some exponential growth into this formula. 🚀
- 📈 Things grow 🚀 exponentially when they go 🦠 viral and 🗣️ people share 💌 things with their 🧑‍🤝‍🧑 friends and 👨‍👩‍👧‍👦 family.
- 💰 Making that happen is the subject keeping many would-be 🤑 billionaires 💻 busy at 🏢 startups.
- 📚 But there are plenty of books on the subject…

🐦 Tweet

Here’s the catchiest tweet I could workshop with AI assistance. 🤞 Let’s see what happens!

🏎️ Supercharge your 🎓 learning with my 🕊️ FREE Obsidian 🤖 AI 📖 book report template! Generate quality, emojified 🪄📄 reports in ⚡ seconds, powered by Google's Gemini API. See examples like 👑 How to Stand Up to a Dictator: ⚔️ The Fight for Our Future.https://t.co/tPwHNOd2lk
— Bryan Grounds (@bagrounds) April 14, 2025

bagrounds.org

Table of Contents

2025-04-13 | 💪 Empower 👥 People

📚 Books

🤖🤖📖 An AI Book Report Template

🤪 Emojify Template

🪞 Reflections

🐦 Tweet

Graph View

Backlinks