Overview
ARC-AGI-3 Docs
Content and agent signals
For agents (AI-generated): ARC-AGI-3 Docs: an interactive reasoning benchmark for AI agents in unseen environments, covering API keys, game schema, actions, scoring methodology, scorecard
arc-agi-3ai benchmarkinteractive reasoningagent evaluationapi keys
Files
Publisher files and the directory’s indexed copy.
Read the cached llms.txt preview
Their llms.txt
The beginning of the file, from our validated copy saved 2026-11-03T04:59:17.041Z. Cached copy · Source on arcprize.org
# ARC-AGI-3 Docs
## Docs
- [ARC-AGI-3 Quickstart](https://docs.arcprize.org/index.md): ARC-AGI-3 is an Interactive Reasoning Benchmark designed to measure an AI Agent's ability to generalize in novel, unseen environments.
- [API Keys](https://docs.arcprize.org/api-keys.md): How to get and use your ARC-AGI-3 API key
- [Game Schema](https://docs.arcprize.org/game-schema.md): Structure and format of ARC-AGI-3 game environments
- [Available Games](https://docs.arcprize.org/available-games.md): List of ARC-AGI-3 games and how to discover them
- [Actions](https://docs.arcprize.org/actions.md): Input interface for ARC-AGI-3 games
- [ARC-AGI-3 Scoring Methodology](https://docs.arcprize.org/methodology.md): How ARC-AGI-3 scoring works
- [Scorecards](https://docs.arcprize.org/scorecards.md): Keeping track of agent performance
- [Recordings & Replays](https://docs.arcprize.org/recordings.md): Viewing your agent's gameplay
- [Agents Quickstart](https://docs.arcprize.org/agents-quickstart.md): Build AI agents for ARC-AGI-3 environments
- [LLM Agents](https://docs.arcprize.org/llm_agents.md): Language model agents for ARC-AGI-3.
- [Benchmarking Tooling (BETA)](https://docs.arcprize.org/benchmarking-agent.md): Run repeatable agent evaluations.
- [AgentOps Template](https://docs.arcprize.org/partner_templates/agentops.md): AgentOps Template
- [HuggingFace Template](https://docs.arcprize.org/partner_templates/huggingface.md): HuggingFace Template
- [Anthropic Template](https://docs.arcprize.org/partner_templates/anthropic.md): Anthropic Template
- [LangChain Template](https://docs.arcprize.org/partner_templates/langchain.md): LangChain Template
- [GPT OSS on Kaggle](https://docs.arcprize.org/partner_templates/gpt-oss-kaggle.md): Run a GPT-OSS-120B agent against ARC-AGI-3 on Kaggle
- [Create Your Own Agent](https://docs.arcprize.org/create-agent.md): Build a custom agent for ARC-AGI-3 games
- [Local vs Online](https://docs.arcprize.org/local-vs-online.md): Playing games locally with the engine vs online via the API
- [Full Play Test](https://docs.arcprize.org/full-play-test.md): What happens under the hood when running an ARC-AGI-3 game
- [Swarms](https://docs.arcprize.org/swarms.md): Orchestrate agents across multiple games.
- [ARC Prize 2026](https://docs.arcprize.org/arc-prize-2026.md): Local dev starter kit for the ARC Prize 2026 Kaggle competition.
- [Learn more about ARC-AGI-3](https://docs.arcprize.org/arc-agi-3.md): Learn more about the ARC Prize and ARC-AGI-3
- [Vocabulary](https://docs.arcprize.org/vocabulary.md): Key terms used in ARC-AGI-3 documentation
- [Feature Requests & Bugs](https://docs.arcprize.org/feature-requests.md): Report issues and request features
- [Contributing](https://docs.arcprize.org/contributing.md): Help build and maintain ARC-AGI-3 Agents.
- [ARC-AGI Toolkit Changelog](https://docs.arcprize.org/changelog.md): Updates and announcements
- [ARC-AGI Toolkit Quickstart](https://docs.arcprize.org/toolkit/overview.md): Getting starte
…
Published headings
2 indexed headings in this publisher record.
Headings describe indexed content. They do not establish a working integration or endorsement.
Agent access
Review the publisher’s own access terms before automated use.
Machine-readable: GET /v1/resolve?q=arcprize.org · GET /v1/domain/arcprize.org · this page as Accept: application/json or text/markdown · MCP lookup_domain
Automatically discovered from public files. Inclusion does not imply endorsement or publisher participation.
Indexing evidence
Authority, classification and country evidence
Own or manage this website?
If you own or manage arcprize.org, you can update its listing and review file health, discovery activity and score history.
Manage this listing — optional
What verification unlocks
See how your site is discovered—and control what agents see.
- See your website in the map
Review your indexed topics, nearby publishers sharing headings, measured listing exposure and map snapshots in a private report for your website.
- Control what agents see
Edit your listing description, tags, public contact details and additional llms.txt paths. Publish your terms for automated access, attribution and pricing.
- Make your profile eligible for search
A valid, listed, unblocked, verified profile can appear in our publisher sitemap and become eligible for search-engine visibility. Search engines decide what to index and rank.
- Appear in our llms-full.txt
Eligible verified publishers are listed by name, with their file links and summaries, in the service overview agents can read.
Free account required for verification and live checks. No backlink or payment required. Public discovery remains available without verification.
Let your coding agent verify it
If your coding agent is authorized to manage this website, paste these instructions into it.
Claim my listing on Agent Web Map for arcprize.org:
1. POST https://agentwebmap.com/v1/sites/arcprize.org/claims and keep claim_id and claim_secret private.
2. Publish the verification line "AgentWebMap verification: <token>" (token from step 1) in https://arcprize.org/llms.txt, or use the DNS, meta-tag or well-known method the response describes. This only verifies the listing; your website and domain remain yours.
3. POST https://agentwebmap.com/v1/sites/arcprize.org/verify with {"claim_id": "...", "claim_secret": "..."}.
4. Save the api_key from the answer; it manages the listing (PATCH https://agentwebmap.com/v1/sites/arcprize.org).
MCP equivalent: tools begin_ownership_claim, then verify_domain, at https://agentwebmap.com/mcp. Docs: https://agentwebmap.com/docs#verifyCorrections or removal requests: contact the directory.