Skip to content

Instantly share code, notes, and snippets.

@Sultan-papagani

Sultan-papagani/main.js

Created November 28, 2025 20:32
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save Sultan-papagani/0ce267d278f5ab3b0b6fd9991a4ad4ae to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save Sultan-papagani/0ce267d278f5ab3b0b6fd9991a4ad4ae to your computer and use it in GitHub Desktop.
Download ZIP
Reddit Post Summarizer with Ollama & Model Dropdown (Tampermonkey)
Raw
main.js
// ==UserScript==
// @name Reddit Post Summarizer with Ollama & Model Dropdown
// @namespace http://tampermonkey.net/
// @version 0.5
// @description Summarize Reddit posts with Ollama, select model from dropdown
// @match https://www.reddit.com/*
// @grant GM.xmlHttpRequest
// @connect 127.0.0.1
// @connect 127.0.0.1:11434
// ==/UserScript==
(function() {
'use strict';
const OLLAMA_URL = 'http://127.0.0.1:11434/api/generate';
const TAGS_URL = 'http://127.0.0.1:11434/api/tags';
let selectedModel = localStorage.getItem('ollama_model') || 'granite3.1-moe';
let availableModels = [];
function fetchModels(callback) {
GM.xmlHttpRequest({
method: 'GET',
url: TAGS_URL,
onload: function(res) {
if (res.status >= 200 && res.status < 300) {
try {
const data = JSON.parse(res.responseText);
availableModels = data.models.map(m => m.name);
if (!availableModels.includes(selectedModel)) selectedModel = availableModels[0];
localStorage.setItem('ollama_model', selectedModel);
callback();
} catch (e) {
console.error('Failed to parse models', e);
}
}
},
onerror: function(err) {
console.error('Failed to fetch models', err);
}
});
}
function collectText(container) {
const paragraphs = Array.from(container.querySelectorAll('p, li'));
return paragraphs.map(p => p.innerText.trim()).join('\n');
}
function summarizePost(container) {
appendSummary(container, null); // show loading
const textToSummarize = collectText(container);
if (!textToSummarize) return;
const payload = {
model: selectedModel,
//summarize the following text very shortly:
prompt: " Very shortly summarize the following text: " + textToSummarize,
stream: false
};
GM.xmlHttpRequest({
method: 'POST',
url: OLLAMA_URL,
headers: { 'Content-Type': 'application/json' },
data: JSON.stringify(payload),
onload: function(res) {
if (res.status >= 200 && res.status < 300) {
try {
const obj = JSON.parse(res.responseText);
let summary = obj?.response || 'No response field found';
// Format Markdown bold and line breaks
summary = summary.replace(/\*\*(.*?)\*\*/g, '<strong>$1</strong>').replace(/\n/g, '<br>');
appendSummary(container, summary);
} catch (e) {
console.error('Failed to parse Ollama response:', e, res.responseText);
appendSummary(container, 'Ollama is not running');
}
} else {
console.error('Ollama error:', res.status, res.statusText);
appendSummary(container, 'Ollama is not running');
}
},
onerror: function(err) {
console.error('Ollama request failed:', err);
appendSummary(container, 'Ollama is not running');
}
});
}
function appendSummary(container, summary = null) {
// Check for existing summary outside the container
let summaryEl = container.nextElementSibling;
if (!summaryEl || !summaryEl.classList.contains('tm-ollama-summary')) {
summaryEl = document.createElement('div');
summaryEl.className = 'tm-ollama-summary';
summaryEl.style.backgroundColor = '#f0f0f0';
summaryEl.style.padding = '10px';
summaryEl.style.borderRadius = '8px';
summaryEl.style.border = '1px solid #ccc';
summaryEl.style.marginTop = '10px';
summaryEl.style.whiteSpace = 'pre-wrap';
summaryEl.style.position = 'relative';
summaryEl.style.width = '100%';
summaryEl.style.clear = 'both';
// Header with dropdown
const header = document.createElement('div');
header.style.display = 'flex';
header.style.alignItems = 'center';
header.style.justifyContent = 'space-between';
header.style.marginBottom = '5px';
const title = document.createElement('span');
title.innerHTML = '✨ <strong>Usefull AI Summarization</strong>';
title.style.fontSize = '14px';
const select = document.createElement('select');
select.style.fontSize = '10px';
select.style.padding = '2px';
select.style.height = '26px';
select.style.minWidth = '60px';
select.style.maxWidth = '150px';
select.style.overflow = 'hidden';
select.style.textOverflow = 'ellipsis';
select.style.whiteSpace = 'nowrap';
availableModels.forEach(model => {
const option = document.createElement('option');
option.value = model;
option.textContent = model;
if (model === selectedModel) option.selected = true;
select.appendChild(option);
});
select.addEventListener('change', () => {
selectedModel = select.value;
localStorage.setItem('ollama_model', selectedModel);
content.innerHTML = 'Loading...';
summarizePost(container);
});
header.appendChild(title);
header.appendChild(select);
// Content div
const content = document.createElement('div');
content.style.fontSize = '13px';
content.innerHTML = summary ? summary : 'Loading...';
summaryEl.appendChild(header);
summaryEl.appendChild(content);
// Insert summary **after the container** (as a sibling)
container.insertAdjacentElement('afterend', summaryEl);
} else {
const content = summaryEl.querySelector('div:nth-child(2)');
content.innerHTML = summary ? summary : 'Loading...';
}
}
function init() {
// Only run on a post page
if (!window.location.pathname.includes('/comments/')) return;
fetchModels(() => {
const observer = new MutationObserver(() => {
const postContainers = document.querySelectorAll('shreddit-post-text-body');
postContainers.forEach(container => {
// Skip if already summarized
if (container.dataset.summarized) return;
// Skip if container is inside the feed (main page)
if (container.closest('shreddit-feed')) return;
container.dataset.summarized = 'true';
summarizePost(container);
});
});
observer.observe(document.body, { childList: true, subtree: true });
});
}
init();
})();
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment