Aptos introduces AI tools for conversational development on Move

Aptos rolled out a suite of AI developer tooling centered on an Aptos Model Context Protocol (MCP) server, updated October 9, 2026, that lets AI coding assistants call Aptos APIs and query on-chain data in natural language and install via a single npx command, with support for Claude Code, Cursor, OpenAI Codex and GitHub Copilot. The release also includes standardized LLMs.txt feeds in full and condensed versions and Aptos Agent Skills, updated September 14, 2026, covering secure Move V2 contract writing, security audits and test generation with 100% coverage. The tooling builds on Aptos's 2024 partnership with FLock.io and aims to shorten the on-ramp for the niche Move language, originally developed for Facebook's Diem project.

Aptos introduces AI tools for conversational development on Move A new MCP server, LLMs.txt feeds and agent skills let AI coding assistants learn Move straight from the documentation Aptos https://cryptobriefing.com/markets/aptos/ wants your AI coding assistant to stop guessing and start reading. The blockchain has rolled out a suite of AI tooling that lets coding assistants learn the Move programming language straight from its documentation, with no manual copy-pasting required. That matters because Move is a niche language. AI models trained mostly on JavaScript and Python have historically treated it like a foreign film without subtitles. What Aptos actually shipped The centerpiece is the Aptos Model Context Protocol MCP server , detailed on the Aptos documentation site and updated on October 9, 2026. The server lets AI tools call Aptos APIs and query on-chain data using natural language. A developer can describe what they want in plain English, and the assistant handles the API plumbing. The server also supports coding and project scaffolding. Installation runs through a single npx command. According to the documentation, the MCP server works with several of the most widely used AI coding tools: - Claude Code - Cursor - OpenAI https://cryptobriefing.com/markets/openai/ Codex - GitHub Copilot LLMs.txt and agent skills The second piece is a set of standardized LLMs.txt feeds. These turn Aptos documentation into machine-readable text that large language models can digest cleanly. The feeds come in two flavors: a full version and a condensed one. AI, tech, and the markets they move—in one daily briefing. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. The third piece is Aptos Agent Skills , which were updated as of September 14, 2026. Per the documentation, the skills cover writing secure Move V2 contracts, running rigorous security audits and generating tests with 100% coverage. Why Move needed the help Move was originally developed for the Diem project at Facebook. When that effort wound down, the language found a second life on Aptos. The language is built around resource safety and formal verification. Resource safety means digital assets are treated like physical objects: they cannot be accidentally copied or quietly deleted. Formal verification means code can be mathematically checked to confirm it behaves as intended. This is not Aptos’s first attempt at bridging AI and Move. In 2024, the project partnered with FLock.io to refine how AI models interact with the language. The new tooling builds on that groundwork. What this means for builders and the ecosystem For developers, the practical change is a shorter on-ramp. Instead of feeding documentation into a chatbot by hand, a builder can point a familiar assistant at Aptos and have it query live chain data, scaffold a project and draft contracts. Smart contract bugs are expensive and permanent once deployed. Tooling that pushes AI toward secure Move V2 patterns, audits and full test coverage targets the exact failure points that have historically burned crypto projects. There is a reasonable caveat. AI-generated code is only as good as the review it receives, and an assistant that writes contracts faster can also write mistakes faster. Agent skills focused on audits and testing appear designed to blunt that risk, but they do not replace human scrutiny. Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy https://cryptobriefing.com/editorial-policy/ .