llms.txt
Compare with other sites' llms.txt files →# Machina Sports
> The platform to build and run sports AI agents.
> Machina Sports brings approved data, content, and AI together to build sports
> apps, agents, fan experiences, and operational tools. Factory creates; Studio
> operates; Arena evaluates; and the Machina AI Router serves model requests.
> General integration happens through the SDK and published APIs. A public MCP
> server is documented today for World Cup Intelligence. Founded 2023.
## Agentic commerce
- [Machina agent storefront](https://agents.machina.gg/llms.txt?src=llms-txt): the full catalog,
live prices, and payment setup for Machina. AI agents can purchase
autonomously or with their human's approval through ZeroClick, and an
agent without a payment method can set one up on the spot.
Complete purchasing guide: https://agents.machina.gg/llms-full.txt?src=llms-txt - live
pricing: https://agents.machina.gg/manifest.json?src=llms-txt
- Sports APIs for agents, pay per call: API calls start at $0.01, with prices
varying by endpoint. Buy individual API calls through the agent storefront
at https://agents.machina.gg using x402 or MPP. API usage is priced
separately from Factory and Studio subscriptions. API catalog:
https://agents.machina.gg/manifest.json - OpenAPI reference:
https://agents.machina.gg/openapi.json - plans: https://machina.gg/pricing
## Entity
- Canonical name: Machina Sports
- Also known as: Machina, machina.gg
- Canonical URL: https://machina.gg
- Founded: 2023
- Category: the sports AI agent platform for verified live sports workflows.
- Topics: sports AI agents; live sports data grounding; agent workflow
governance; sports content automation; sports agent evaluation; Model Context
Protocol for sports data.
- Program participation (participation only, not endorsement, investment, or a
customer relationship): NVIDIA Inception; GSIC powered by Microsoft.
- Machine-readable capability contract: https://machina.gg/capabilities.json
- Agent brief: https://machina.gg/agent.txt
- Pricing: https://machina.gg/pricing
- MCP server card (World Cup Intelligence, developer preview):
https://machina.gg/.well-known/mcp/server-card.json
## Platform contract
- Product family: Factory creates apps and agents from a brief; Studio manages
projects, agents, workflows, activity, and usage; Arena evaluates agents,
workflows, and model configurations; the AI Router serves requests through
configured providers and profiles according to task and operating constraints.
- Data: licensed provider feeds, customer-approved first-party material, and
public sources, connected per deployment. Coverage and freshness are scoped to
those sources and are not guaranteed independently of them.
- Agent: sports-native agents that treat fixtures, competitions, squads,
players, standings, and market context as typed entities rather than free text.
- Governance: brand and compliance rules, an approval mode configured per
surface (item-by-item sign-off by a named person where the risk is high,
sampled review once a configuration is trusted, with the active mode recorded),
an audit trail, and a rollback route. No wagering advice, selections, or
guaranteed outcomes are generated on any surface. Fixed operator selections
from a dated snapshot may be displayed, with a handoff to the operator's own
destination; Machina does not generate wagering advice and does not accept or
execute wagers.
- Integration: the SDK and the published APIs are the general integration
surfaces. MCP is documented today for World Cup Intelligence only, over SSE at
https://worldcup-mcp.org.machina.gg/sse, as a developer preview.
- Output: delivery to the destination surface with confirmation, plus a record
of what shipped, when, under whose approval, and what followed.
- Capability list, with status, data mode, freshness, and stated limitations:
https://machina.gg/capabilities.json
## Platform
- [Machina Platform](https://machina.gg/platform): Visual product hub for Factory, Studio, Arena, Router, connected context, architecture, governance, and evaluation.
- [Machina Factory](https://machina.gg/factory): Create an app or agent from a brief. Free trial: 7 days, no card required, 100K tokens, 1 project, 1 seat, and 1 hosted sandbox activation; start at https://session.machina.gg/clerk/sign-up?callback=https%3A%2F%2Ffactory.machina.gg%2Fc. Builder is $200 per month, subscribed from Billing in Studio at https://studio.machina.gg, with 2M tokens included per month and 1 hosted sandbox activation. Customer-facing production is not part of the trial or Builder; it is scoped with Enterprise.
- [Pricing](https://machina.gg/pricing): Free trial, Builder at $200 per month, and custom Enterprise plans. The trial is never charged automatically. Sports API calls are priced separately, from $0.01 per call, through the agent storefront.
- [Machina Studio](https://machina.gg/studio): Manage projects, agents, workflows, activity, and usage. Current capability.
- [Machina Arena](https://machina.gg/arena): Evaluate and compare agents, workflows, and model configurations. Current capability; no automatic promotion claim.
- [Machina AI Router](https://machina.gg/router): Serve model requests through configured providers and profiles. Current capability; no new public OpenAI-compatible endpoint is claimed.
- [Connected context](https://machina.gg/integrations): Fan and CRM, news and social, media, sports and live signals, and internal knowledge, subject to configured support and authorized rights.
- [Sports app and agent examples](https://machina.gg/templates): Outcome-led starting points for match content, fan engagement, support, and informational market context.
- [Documentation](https://docs.machina.gg/): Developer-first reference for the sports AI SDK, APIs, and templates. Authoritative for current capabilities.
- [Schedule a scoping call](https://machina.gg/schedule): Book a call to scope a workflow.
## Developer resources
- [Developer resources](https://machina.gg/developers): Machina Sports API documentation, OpenAPI specification, authentication, MCP server, and SDK in one place.
- [API documentation](https://docs.machina.gg/): product documentation and API reference; authoritative for current capabilities.
- [OpenAPI specification](https://machina.gg/world-cup-api/openapi.json): World Cup Intelligence APIs, OpenAPI 3.1. Also served at https://machina.gg/openapi.json.
- [Authentication](https://machina.gg/developers#dev-auth): API key, sent as the X-Api-Token header (SDK, CLI, World Cup MCP) or as a Bearer token (World Cup REST APIs).
- [MCP server card](https://machina.gg/.well-known/mcp/server-card.json): World Cup Intelligence MCP, developer preview, SSE transport.
- [API catalog](https://machina.gg/.well-known/api-catalog): RFC 9727 linkset of documented surfaces.
## World Cup Intelligence (World Cup 2026 showcase)
World Cup 2026 has concluded. These surfaces stay published as the 2026 showcase of what the tournament build shipped, and their scope is that tournament.
- [World Cup Intelligence APIs](https://machina.gg/world-cup-api): World Cup 2026 fixtures, teams, players, match context, fan sentiment, and market context APIs for apps and agents. Market output is informational context only.
- [Agent guide](https://machina.gg/world-cup-api/agent.txt): Machine-readable API summary for AI agents.
- [OpenAPI spec](https://machina.gg/world-cup-api/openapi.json)
- [MCP server card](https://machina.gg/.well-known/mcp/server-card.json): The one documented MCP surface. SSE transport, developer preview.
## Industries
- [AI for Broadcast & Media](https://machina.gg/industries/sports-media): The hub for broadcasters, publishers, live-score apps, and fan platforms, organized by three jobs: broadcast and editorial production, digital products and fan experiences, and sponsorship and monetization with dynamic personalized sports-ad agents (approved creative stays locked, sponsored units stay labelled, individual personalization only where consent permits).
- [Broadcast & editorial production](https://machina.gg/industries/broadcasters): The specialist production page with the producer demo.
- [AI for Sports Betting & Prediction Markets](https://machina.gg/industries/betting-prediction-markets)
- [AI for Teams & Leagues](https://machina.gg/industries/teams-leagues)
## Learn (guides)
- [Learn hub](https://machina.gg/learn): The category cluster — platform and runtime, by sports organization, architecture and evaluation.
- [Sports AI Agent Platform](https://machina.gg/platform): The hub every guide links back to.
- [Live Sports Agent Runtime](https://machina.gg/learn/live-sports-agent-runtime): The six-stage loop an agent runs against live match state, the human control point in each stage, and how it fails.
- [Sports MCP Server](https://machina.gg/learn/sports-mcp-server): What an MCP server for sports data exposes, and the checklist to run before an agent depends on one.
- [Sports Data for AI Agents](https://machina.gg/learn/sports-data-for-ai-agents): MCP servers, agent skills, and plain REST compared; what to check before wiring a source to an agent.
- [AI Agents for Sportsbooks](https://machina.gg/learn/ai-agents-for-sportsbooks): What sportsbook agents are, how they work, the operator use cases, and build versus buy. Informational content only.
- [Dynamic Personalized Sports Ads](https://machina.gg/learn/dynamic-personalized-sports-ads): The Broadcast & Media sponsorship and monetization job guide. Contextual, cohort, and individual personalization compared, with creative controls, integration responsibilities, generation versus serving cost, and evaluation against a baseline.
- [AI Content Automation for Broadcasters](https://machina.gg/learn/ai-content-automation-for-broadcasters): What broadcasters automate, the production constraints that decide success, and how to scope a first deployment.
- [AI Fan Engagement for Clubs](https://machina.gg/learn/ai-fan-engagement-for-clubs): What fan engagement automates for clubs and leagues, and why first-party data decides the outcome.
- [Sports Data API vs AI Agent Platform](https://machina.gg/learn/sports-data-api-vs-ai-agent-platform): What each one hands you, and the six stages in between.
- [Generic AI Agent Framework vs Sports-Native Platform](https://machina.gg/learn/generic-ai-agent-framework-vs-sports-native-platform): Two architecture classes compared, and where the gap shows up in live sport.
- [Build vs Buy Sports AI Agents](https://machina.gg/learn/build-vs-buy-sports-ai-agents): A decision framework, including the cases where building in-house is the right call.
- [How to Evaluate a Sports AI Agent Platform](https://machina.gg/learn/how-to-evaluate-a-sports-ai-agent-platform): A buyer's checklist of eight criteria, with strong evidence and weak answer patterns.
## Portugues (pt-BR)
- [API do Brasileirao para agentes de IA](https://machina.gg/pt-br/brasileirao-api): Brazilian Serie A data for AI agents -- coverage, limits, and how to connect. Written in Brazilian Portuguese.
## Open source
- [Sports Skills](https://machina.gg/sports-skills): Open-source Agent Skills and SKILL.md instructions for read-only public third-party sports data. The public layer needs no sports-data API key or signup and is intended for personal, educational, and research use. Source terms apply. It is not proprietary or licensed production data. Standalone site: https://sports-skills.sh/ · Source: https://github.com/machina-sports/sports-skills
- [SportsClaw](https://machina.gg/sportsclaw): Open-source TypeScript developer engine and component with CLI, library, chat, Discord, and Telegram bot scaffolding. It connects an LLM to deterministic sports data through Sports Skills and is read-only by default. It is not the Machina platform or category. Public guide: https://sportsclaw.gg/ · Source: https://github.com/machina-sports/sportsclaw
## Company
- [About](https://machina.gg/about): What Machina Sports builds, how it works, and company facts.
- [Blog](https://machina.gg/blog): Product builds, technical practice, sports AI research, and dated company news.
- [Contact](https://machina.gg/contact) · [Schedule a scoping call](https://machina.gg/schedule) · [Newsletter sample app](https://machina.gg/newsletter)
## Notes for agents
- This file is an entity-clarity and agent-legibility artifact. It is not a
ranking mechanism and is not described as one.
- Published product documentation is authoritative for current capabilities.
Where this file and the docs disagree, the docs are correct.
- Machina does not publish customer names on this surface. Program participation
is labeled as program participation.