> For the complete documentation index, see [llms.txt](https://docs.strikebit.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.strikebit.io/introduction/multi-agent-architecture-and-map.md). # Multi-Agent Architecture & MAP StrikeBit’s multi-agent architecture enables the creation of layered, interoperable AI systems where primary agents can deploy and manage specialized subagents, each assigned to handle distinct tasks such as data analysis, user interaction, or logic execution. A core innovation of this design is that agents are built to learn from one another, continuously exchanging insights to improve performance and expand their collective intelligence. This collaborative structure not only enhances scalability and adaptability but also introduces a new standard of composability — where users can evolve their agent ecosystems by upgrading, combining, and monetizing individual modules over time.

Multi-Agent Architecture

## **Our Vision: Multi-Agent Architecture & Modular Agent Protocol (MAP)** The current generation of “AI agents” in Web3 are often isolated LLM wrappers — single-function bots without composability, upgradability, or true ownership. StrikeBit’s MAP addresses this by enabling: * **Composable Architecture**: Agents can plug into each other like smart contracts. * **Onchain Ownership**: Each agent is tokenized (via ERC721 or ERC404) and user-owned. * **Modular Logic**: Agents can be deployed with or upgraded via new logic modules. * **Collaborative Networks**: Agents can delegate, receive, or inherit logic from other agents (multi-agent systems). ### Key Principles of MAP #### 🧠 1. **Modularity** Every agent is composed of **modules** — smart contract components responsible for: * Input/output interpretation * Access control * Onchain triggers (e.g., token gates, staking thresholds) * Execution logic (trade, alert, generate, distribute, etc.) These modules can be swapped or added over time, allowing the agent to evolve. *** #### 🤝 2. **Multi-Agent Collaboration** Agents can operate independently, or as **multi-agent systems**, where: * Subagents specialize in tasks (e.g., one for data scraping, one for trading) * Agents call on one another to perform complex workflows * Developers create **agent stacks** where each layer builds on the last *** #### 🔗 3. **Programmable Ownership** Agents launched on MAP are minted as **ERC721 NFTs** (or ERC404 if applicable). This means: * The creator or DAO can control upgrades * Holders can vote on governance or unlock modules via staking * Usage rights (API access, gated functions) can be token-gated via ownership *** #### 🧠 4. **Deployable by Anyone** Through AgentHub, users can: * Migrate existing agents from Virtuals * Create agents from templates * Customize agent utility (via $HUB or other onchain payments) * Mint the agent as an NFT and list it on the StrikeBit Marketplace --- # 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: ``` GET https://docs.strikebit.io/introduction/multi-agent-architecture-and-map.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.