{"id":619506,"date":"2026-10-03T15:06:16","date_gmt":"2026-10-03T15:06:16","guid":{"rendered":"https:\/\/www.newsbeep.com\/nz\/619506\/"},"modified":"2026-10-03T15:06:16","modified_gmt":"2026-10-03T15:06:16","slug":"chatgpt-6-astra-plays-world-of-warcraft-blind-and-clears-the-orc-starting-zone-in-40-minutes-with-no-deaths-ai-agent-navigates-by-parsing-raw-server-network-packets-and-sql-files","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/nz\/619506\/","title":{"rendered":"ChatGPT-6 Astra plays World of Warcraft ‘blind’ and clears the orc starting zone in 40 minutes with no deaths \u2014 AI agent navigates by parsing raw server network packets and SQL files"},"content":{"rendered":"

OpenAI’s GPT-6 Astra AI model was able to clear the Orc starting area in World of Warcraft (WoW) in 40 minutes with 0 deaths, agent-wow’s developer<\/a> says. The model played the zone without seeing a single rendered frame, relying on network traffic and quest data pulled from the server’s own files. The approach used a single prompt in Codex with agent-wow, an open-source client, to play on a private server. The corresponding YouTube video says that the agent \u201cstarts as a level 1 Orc, completes every quest in the Valley of Trials, and finishes the run in Sen\u2019jin Village.\u201d<\/p>\n

Go deeper with TH Premium: GPUs<\/p>\n

\n

\"Asus\n<\/p>\n

(Image credit: Noctua)<\/p>\n

Agent-wow is an \u201cAzerothCore WoW client designed for autonomous AI agent players,\u201d according to the GitHub repo description. The client does not \u201cdefine any gameplay mechanics such as movement, combat, or in-game interactions\u201d and instead only exposes a \u201cmodule system for agents to build\u201d whatever they need to play. AzerothCore is open-source server software that runs WoW 3.3.5a, \u201cthe final build of Wrath of the Lich King,\u201d and agent-wow talks to it over the game\u2019s network protocol. The client does not connect to any live WoW server and instead can be run on local private servers for experimentation.<\/p>\n

<\/p>\n

The developer ran OpenAI\u2019s flagship model<\/a>, released early last month, with extra high reasoning effort, one below the highest or maximum. The developer chose WoW for its mix of long-term strategy and short-term tactics, with an end goal of filling \u201can entire server with AI agents [to] see if they can clear Icecrown Citadel on heroic difficulty.\u201d<\/p>\n

Latest Videos FromTom’s Hardware<\/p>\n

The agent built a single module to grab 28 types of server messages, by our count, which are kept in memory. A Python script polls those messages to build the agent\u2019s picture of the world and sends messages back to act in it. The developer expected a higher-level implementation, but \u201cIn practice, it was more than capable of working at the protocol layer.\u201d<\/p>\n

GPT-6 Astra Clears the Orc Starting Zone in World of Warcraft | agent-wow – YouTube
\n\"GPT-6<\/p>\n

Watch On <\/a><\/p>\n

For quest data, the agent turned to data mining, pulling quest givers, turn-ins, and spawn points from AzerothCore\u2019s SQL files. The developer says this is \u201carguably similar to how a human might spend hours on Wowhead to research quests,\u201d referring to a popular WoW info site. Going straight to the server\u2019s own files arguably beats that, since they\u2019re the data the server itself runs on, while a fan site\u2019s data may not always match. The project\u2019s public workspace instructions also list AzerothCore\u2019s source code as a resource for agents, though the developer doesn\u2019t say whether this run used them.<\/p>\n

You may like<\/p>\n

Some planning was used in the agent\u2019s approach. According to the developer, it worked through prerequisite quest chains in order, sold junk, equipped upgrades, and trained abilities before the zone\u2019s final cave, and picked up both cave quests at once to complete them together.<\/p>\n

For pathfinding, the agent built a C++ helper that plots a route between a starting point and a destination with AzerothCore\u2019s own navigation mesh files, or mmaps. The Detour pathfinding library finds the route, and the helper returns its waypoints as coordinates, or an error if there\u2019s no complete path. The developer acknowledges that \u201cfrom my research into heuristics-based bots, pathfinding is always one of the main challenges,\u201d but calls the agent\u2019s pathfinding abilities \u201coptimal.\u201d In fact, the agent was able to \u201cexploit map bugs\u201d at locations where collision properties may have been lacking.<\/p>\n

Get Tom’s Hardware’s best news and in-depth reviews, straight to your inbox.<\/p>\n

This is not the first game the model has tackled. Early last month, shortly after its release, it played the game Portal<\/a> with screenshots and knowledge of the player\u2019s position. This WoW run used no screenshots at all and only covered the first zone, while the Portal run went through the entire game over about 24 hours.<\/p>\n

Recently, developers have deployed various techniques to play Pok\u00e9mon Red, such as using a custom small model<\/a> and a Jev decision model harness<\/a> with Claude coaching. Next up for the developer: discovering whether a single agent can reach level 80 completely on its own, and whether multiple agents can team up to complete content through the game\u2019s social features.<\/p>\n

<\/p>\n

\n

\"Google\n<\/p>\n

<\/a><\/p>\n

Follow Tom’s Hardware on Google News<\/a>, or add us as a preferred source<\/a>, to get our latest news, analysis, & reviews in your feeds.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenAI’s GPT-6 Astra AI model was able to clear the Orc starting area in World of Warcraft (WoW)…\n","protected":false},"author":2,"featured_media":619507,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[365,363,364,111,139,69,145],"class_list":["post-619506","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-new-zealand","tag-newzealand","tag-nz","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/619506","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/comments?post=619506"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/619506\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media\/619507"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media?parent=619506"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/categories?post=619506"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/tags?post=619506"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}