> For the complete documentation index, see [llms.txt](https://docs.aloop.icustomer.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aloop.icustomer.ai/concepts/context-and-memory.md). # Context and memory What the system knows and what it has learned: sources, precedence, and why what you say always wins. Everything the system knows lives in one place: the **Growth Brain**, the self-learning context iHarness runs on. Everything you connect accrues into it, and every outcome comes back to it. * **One store, not two.** iHarness's memory and the Growth Brain are the same thing. * **Not a CDP, a data lake, or a warehouse.** It sits on your data infrastructure and does not replace it. Inside it, two kinds of knowing: * **Context** is the current picture of your business: your audience, brand, and competition, your funnel, and anything you have told the system directly. Context can be handed over. * **Memory** is what the system has learned by running: which signals mattered, which moves worked, what your team approves and rejects. Memory has to be earned. ## Where context comes from Context is seeded from public information about your company, then sharpened by your connected tools and by you. Every claim carries its source, and claims that could not be verified are presented as questions rather than stated as facts. ## Which source wins When sources disagree, precedence is fixed and human-first: 1. **What you have declared.** A correction from you outranks everything, permanently. 2. **Your CRM.** Your system of record beats anything inferred. 3. **Observed signals.** What actually happened on your site and channels. 4. **Enriched and public sources.** Useful, and the first to yield. Corrections are versioned: when you change the picture, decisions made afterward reference the picture that was in force, which is what keeps [traces](/concepts/decisions-and-traces.md) honest over time. ## What memory holds * Audience behavior, and what precedes conversion. * Which messages and moves worked. * Your team's working patterns: what gets approved, changed, and rejected. * The resolved results of past decisions. A decision only becomes memory when its outcome is in. The system learns from what happened, not from what it hoped. You can also **pin facts** you want the system to always remember, and correct anything it has learned. Closed-loop, self-learning optimization over memory is a Team and Enterprise capability. ## Why it compounds Context makes the first decision possible. Memory makes the thousandth one better. Evidence earned in your workspace outranks generic benchmarks the moment it exists, which is why the system sharpens with use instead of going stale. See [The compounding loop](/measurement-and-outcomes/compounding-loop.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.aloop.icustomer.ai/concepts/context-and-memory.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.