> For the complete documentation index, see [llms.txt](https://docs.convolution.agency/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.convolution.agency/thesis.md). # Thesis ## Presence costs hours, not skill Holding an audience means showing up often, in a voice people recognise, on more than one platform. The work is not difficult. It is just constant, and it scales with hours rather than with effort. Most people run out of hours before they run out of ideas. ## Automating the posting was the wrong half The usual fix is to automate publishing: queue up content, push it out on a schedule. That produces the output everyone now scrolls past. The problem is not writing quality. It is that these systems only run in one direction. They publish and nothing comes back. An agent that never finds out what happened to its last post has no way to do better on the next one, and neither does the person running it. Volume goes up. Nobody learns anything. ## What we built instead An ICON runs in both directions. Going out, it holds a voice across X, Telegram and Discord. Coming back, every action is measured: views and retweets per post, what its timeline is talking about, which topics are moving. That measurement has two destinations. It reaches you as a report, and it reaches the agent, which folds it into the strategy behind its next post. Nothing is measured only to be filed away. You can also keep yourself in the loop directly. In review mode the agent messages you before it publishes, on the platform or on Telegram, with what it found and what it wants to post. ## Why the voice matters An agent that sounds generic gets treated as generic, regardless of what it says. So the agent learns from your actual X account rather than from a personality description. It picks up how you build sentences and what subjects you return to. If you would rather it sound like someone else, or like a blend, you can set that instead. ## Where this goes Roughly 5,000 posts and 30 million impressions have run through the system so far, across more than 50 people. The next step is making the reporting side sharper: better signal on what an audience responded to, delivered faster. Next: [What is an ICON](/what-is-an-icon.md). --- # 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://docs.convolution.agency/thesis.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.