> For the complete documentation index, see [llms.txt](https://help.contouron.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.contouron.com/getting-started/understanding-context.md). # Understanding Context > For the complete documentation index, see [llms.txt](https://docs.contouron.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.contouron.com/start/sources.md). ## Understanding Context Context is the raw material you use to generate social media post drafts. Add a YouTube video, article, or PDF as context, then Contouron uses AI to turn that content into post drafts for each platform. Context and drafts are different: * **Context** = raw materials (videos, articles, PDFs, text) stored locally in your browser * **Drafts** = social media posts generated from your context, saved to Contouron's database Here's a step-by-step guide to using context effectively in Contouron: ### Context Supported Click "+ Add Context" in the top-left corner of the [AI Agent](https://my.contouron.com/agent). To generate rich well-researched posts, focus on the QUALITY of each context item! Here is all the context currently supported by Contouron: * **Text:** Copy and paste any relevant text from articles, notes, or documents. * **Articles & Websites:** Input the website URL and Contouron will extract the text content. * **YouTube & TikTok:** Add a link, and Contouron will extract the transcript. The video must have **English captions (subtitles)** for transcript extraction to work. Transcript extraction only reads English, so a non-English video, or a video with no captions, fails with an "error fetching" message. For those videos, copy-paste the content as Text instead. Note: Facebook Reel and Instagram URLs are not supported as video context — use Text or Articles instead. * **Audio, Podcasts, & Meeting Recordings:** Upload audio files to extract useful content. * **PDFs:** Upload PDF documents, ebooks, research papers, case studies, etc. * **Perplexity AI:** Research any topic by using Perplexity's AI-powered real-time web search. > :bulb: The quality of your context directly affects your final output — use reliable, well-structured context for the best results. Garbage In, Garbage Out. > :warning: Context is stored locally in your browser and is NOT saved across devices. If you switch from laptop to phone, your context will not transfer over. *** ### Toggling Context On/Off Once you've added context, toggle each item on or off based on relevance. Only the toggled-on items will be used when you generate a new post or re-generate an existing post. 🚀 Pro Tip: If you're overwhelmed adding context, start by adding ONE high-quality item. I personally don't often use more than 2-3 context items. A SINGLE high-quality piece of context can be used to generate many diverse posts. *** ### Saving Sources to Your Sources Library By default, context is stored locally in your browser and does not carry across devices. To keep a source across sessions, save it to your Sources library: 1. Open the [AI Agent](https://my.contouron.com/agent). 2. In the left sidebar, click the source you want to keep. 3. Check the "Save Source?" box. 4. The source now appears in your [Sources library](https://my.contouron.com/sources). To save more sources, repeat these steps for each one. The Sources page has no separate "add" button — you save sources from the AI Agent sidebar. *** ### Workaround for Paywalled or Inaccessible Websites Contouron can only extract the visible text from websites that are publicly accessible. Paid articles are not supported. Here's a workaround if adding the website URL does NOT work: * Manually copy and paste the website's text. * Add it as "Text" context. *** ### Research Any Topic with Perplexity AI If you're not sure what context to use, let Perplexity do the research for you! :point\_right: Add "Perplexity" as context. 🔍 Type any prompt (e.g. "latest AI trends") 📜 Contouron uses Perplexity AI to research the web. *** ### Custom Instructions to Pre-process Each Context Item "Custom Instructions" allow you to apply a prompt to each context item before generating your final content. When using multiple context items, custom instructions helps AI prioritize key information – ensuring only the most impactful data, statistics, or insights are used to write your posts. Example: if you're using a long Youtube video as context, a custom instruction like *"Extract the top three takeaways"* ensures the post remains concise and valuable. If you have multiple context items, I highly recommend using custom instructions. They lead to **higher-quality content by instructing AI to focus on what's important.** *** ### Examples of Custom Instructions Here are some examples of custom instructions. | Purpose | Custom Instruction | | ------------------------------ | ---------------------------------------------------------------------------------------------- | | Extract Main Content Only | Ignore sidebars, ads, and navigation menus; extract only the core article or video transcript. | | Summarize Key Points | Generate a concise bullet-point summary of the most important insights from the context. | | Highlight Actionable Tips | Identify and list any actionable tips or strategies mentioned in the content. | | Emphasize Quotes | Extract and highlight impactful quotes, ensuring they're clearly formatted and attributed | | Conversational Rewrite | Rephrase the source material in a friendly, conversational tone suitable for social media. | | Create a Step-by-Step Guide | Break down the content into a simple, numbered process or checklist. | | Focus on Data-Driven Insights | Prioritize statistical data and research findings. | | Format for Readability | Reformat the content into short, digestible paragraphs with subheadings where applicable. | | Identify Trends and Patterns | Highlight any emerging trends or recurring themes mentioned in the context. | | Generate FAQs | Based on the content, produce a list of potential FAQs with brief answers. | | Extract Problem/Solution Pairs | Identify any problem statements and their corresponding solutions, listing them side-by-side. | | Focus on Expert Opinions | Extract and emphasize expert opinions, including any credentials or references. | | Emphasize Urgency | Highlight any points that suggest urgency or immediate action, framing them as timely advice. | | Data Exctraction | Extract only the most shocking facts and numerical data. | --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://help.contouron.com/getting-started/understanding-context.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.