> For the complete documentation index, see [llms.txt](https://dopameme.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://dopameme.gitbook.io/docs/protocol/memory-and-learning.md). # Memory and Learning Dopameme's learning loop depends on comparing new evidence against prior packets. When the system sees a token again, it should not treat the token as brand new. It should ask what changed, whether the earlier read held up, and whether the evidence improved or weakened. ## Why Memory Matters * It prevents repeated shallow reads * It creates accountability for prior calls * It lets audits improve future packet quality * It turns scans into a compounding knowledge base The protocol becomes more useful when evidence is preserved and reviewed. ## How Memory Improves The Read Memory gives Dopameme a baseline. If a token was scanned yesterday, the next read should not only describe the current chart. It should explain whether the earlier thesis still makes sense. If the operator read said acceptance above a level mattered, the next scan should know whether acceptance happened. If holder concentration was dangerous, the next packet should know whether that risk improved or got worse. This creates a more useful terminal because the user gets continuity. They are not forced to rebuild the entire context from scratch every time a token moves. ## What Gets Remembered Dopameme memory can include: * Packet evidence from scans * Operator reads and consequence maps * Technical levels and whether they later mattered * Holder context and concentration changes * Tracked-wallet behavior over time * Suppression reasons and whether those filters were useful * Audit decisions and repeated quality issues Memory is only useful when it is organized. Raw history by itself becomes noise. Dopameme's job is to compress history into the parts that improve the next read. ## Learning Is Not Magic Dopameme does not become useful by collecting data alone. The value comes from organizing observations, comparing them, auditing them, and making the next read clearer. Packet memory is the foundation for that process. The protocol should get better because every important scan leaves behind evidence. Over time, that evidence helps Dopameme understand which wallets deserve more trust, which market filters are protecting users, which technical levels matter most, and which operator-read patterns need improvement. --- # 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://dopameme.gitbook.io/docs/protocol/memory-and-learning.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.