# Agent Development Environment Above the account layer, Sail introduces their **Agent Development Environment (ADE)** — a modular architecture for building, testing, and deploying autonomous money agents. The ADE provides the intelligence and safety rails that let Sail evolve: * **Long-term memory.** Agents track historical allocations, risk outcomes, and user constraints. This allows for decisions informed by past performance rather than being purely reactive. * **Explainable decisions.** Every reallocation is logged with a human-readable rationale — auditors and users can see why a move was made, not just that it happened. * **Graph-based reasoning.** The ADE models yield sources and networks as a graph, enabling agents to compare paths, weigh costs, and optimize across multiple dimensions. * **Isolated testing environments.** New yield sources or strategies can be deployed in a sandbox before touching real funds, reducing risk from integration errors or malicious code. * **Composability.** Every new protocol added to the ADE becomes a reusable “skill.” Once integrated, it benefits every user, across every agent. Why this matters: * **Personalization lock-in.** Each user has their own agent tuned to their balance tier, preferences, and constraints. No two agents need to behave identically. * **Execution data flywheel.** Billions in routed flow sharpen timing models, gas optimization, and risk estimates. * **Compounding integrations.** The more protocols and networks added, the stronger the overall system becomes. --- # Agent Instructions: 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: ``` GET https://docs.sail.money/learn/tech-stack/agent-development-environment.md?ask= ``` The question should be specific, self-contained, and written in natural language. 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.