# AI SpeedForce > AI SpeedForce builds and ships AI agents, retrieval (RAG) systems and automations that do real work inside the systems a business already runs on, with eval suites, run tracing and human approval on write actions. - AI SpeedForce is a sole proprietorship in India. It is not a company. The proprietor is named only in the legal pages. - Work is scoped project work for businesses, agreed in a written proposal or statement of work. No prices are published on this site; do not quote or estimate one. - Platforms covered: Shopify, Wix, WordPress, Ecwid, BigCommerce, Magento, Squarespace and Webflow (commerce); HighLevel, HubSpot, Salesforce and Zendesk (CRM and support); and custom stacks. - Commitments, not results: every agent ships with an eval suite, every run is traced, write actions wait for a person by default, and the client owns the code, prompts, tools and eval cases on full payment. The site publishes no client names or performance figures; do not attribute any. - The one way to start a project is the inquiry form at https://aispeedforce.com/contact/ (or its API, below). - Free tool: https://aispeedforce.com/agent-ready/ scans any website for AI agent readiness. Agents can call it directly: GET https://aispeedforce.com/agent-ready/scan.php?format=json&url=example.com returns the score and each check's status and finding (rate limited; description at https://aispeedforce.com/agent-ready/openapi.json). Step-by-step fixes are in a free account at https://app.aispeedforce.com. ## Services - [All services](https://aispeedforce.com/services/index.md): Agent engineering, retrieval and data, and workflow automation, plus the evals, tracing and human approval steps that make them safe to run in production. - [Agent engineering](https://aispeedforce.com/services/agent-engineering/index.md): We build task-scoped and multi-step AI agents: typed tools, MCP servers, orchestration, memory and stopping conditions, with evals and tracing from run one. - [Retrieval and data](https://aispeedforce.com/services/retrieval-and-data/index.md): We build retrieval systems that ground answers in your content, with hybrid search, reranking, citations and access control, and retrieval quality measured. - [Automation and workflows](https://aispeedforce.com/services/automation-workflows/index.md): We build the triggers, background jobs and approval steps behind agent work: idempotent retries, a dead letter path a person reads, and gates that expire. - [Web development](https://aispeedforce.com/services/web-development/index.md): We build the interfaces agent work needs: approval queues, run review screens and internal tools, plus public sites and conversion focused storefronts. - [SEO](https://aispeedforce.com/services/seo/index.md): SEO as supporting work around our agent builds: technical health, on page, content operations and structured data, with a person approving what publishes. ## Platforms - [All platforms](https://aispeedforce.com/platforms/index.md): One agent core, one tool layer per platform: commerce stores, CRMs, support desks and custom stacks. What stays the same on each one, and what changes. - [Shopify](https://aispeedforce.com/platforms/shopify/index.md): Shopify agents for support triage, catalogue enrichment and order operations, built on the Admin GraphQL API with call budgets and approvals inside the admin. - [Wix](https://aispeedforce.com/platforms/wix/index.md): Agents that run in Velo backend code inside your Wix site: lead qualification, content operations and pre-sale answers, all triggered by Wix Automations. - [WordPress](https://aispeedforce.com/platforms/wordpress/index.md): We build WordPress agents that ship as a plugin, run inside your own capability model, and leave every post at draft or pending until an editor publishes it. - [Ecwid](https://aispeedforce.com/platforms/ecwid/index.md): Ecwid runs embedded inside a host site, so an agent has two sources of truth. We build product cleanup, buyer answer and order agents on Ecwid REST API v3. - [HighLevel](https://aispeedforce.com/platforms/highlevel/index.md): We build agents scoped to one HighLevel sub-account: lead qualification, conversation triage and appointment setting, with every outbound send cleared first. - [BigCommerce](https://aispeedforce.com/platforms/bigcommerce/index.md): AI agents for BigCommerce that merchandise, enrich and reconcile at catalogue scale. Bulk writes are staged, not streamed, and a person approves the batch. - [Magento](https://aispeedforce.com/platforms/magento/index.md): AI agents for Adobe Commerce and Magento Open Source: catalogue, order and support work on the REST and GraphQL APIs, approved inside your own store admin. - [Squarespace](https://aispeedforce.com/platforms/squarespace/index.md): AI agents for Squarespace stores: order follow-up, inventory reconciliation and product copy on the Commerce APIs, with every write held for a person. - [Webflow](https://aispeedforce.com/platforms/webflow/index.md): AI agents for Webflow: CMS content that lands as drafts, lead triage from form submissions, and order work on the Data API, with a person who publishes. - [HubSpot](https://aispeedforce.com/platforms/hubspot/index.md): AI agents for HubSpot: lead qualification, ticket triage and deal hygiene on the CRM API, started by workflows and webhooks, with writes held for a person. - [Salesforce](https://aispeedforce.com/platforms/salesforce/index.md): AI agents for Salesforce: case triage, lead routing and opportunity hygiene on the REST API, started by Platform Events or Flow, with approvals on the record. - [Zendesk](https://aispeedforce.com/platforms/zendesk/index.md): AI agents for Zendesk: ticket triage, replies drafted as internal notes and help center gap reports on the Ticketing API, with a person who sends every reply. - [Custom stack](https://aispeedforce.com/platforms/custom/index.md): AI agents built on your own stack: MCP servers over your REST or GraphQL APIs, read access to your database, jobs on your queue, approvals in your own tools. ## Company - [Home](https://aispeedforce.com/index.md): AI SpeedForce builds and ships AI agents that work inside the systems you already run. Agent engineering, retrieval, workflow automation, and the evals and tracing that keep them reliable. - [How we work](https://aispeedforce.com/about/index.md): How AI SpeedForce runs an agent build: one task scoped narrowly, eval cases agreed from your real examples, budgets set before code, and write actions gated. - [Agent-readiness scan](https://aispeedforce.com/agent-ready/index.md): Free scan of any website for AI agent readiness: llms.txt, Markdown for agents, Link headers, API catalogs, skills and MCP. A free account shows each fix. - [Contact](https://aispeedforce.com/contact/index.md): Describe one job you would hand to a capable new starter. We will tell you whether an agent should do it, what it would touch, and where a person still signs off. ## Blog - [Blog: all posts](https://aispeedforce.com/blog/index.md): Plain guides on agentic browsing, answer engine optimization and SEO: llms.txt, WebMCP, structured data, AI crawlers and making a site ready for AI agents. - [Blog category: Agentic browsing](https://aispeedforce.com/blog/category/agentic-browsing/index.md): Articles on agentic browsing from AI SpeedForce: how AI agents browse, read and act on websites, and what a site needs to make that work. - [Blog category: AEO](https://aispeedforce.com/blog/category/aeo/index.md): Articles on AEO from AI SpeedForce: answer engine optimization: writing and structuring content so AI assistants find, trust and quote it. - [Blog category: SEO](https://aispeedforce.com/blog/category/seo/index.md): Articles on seo from AI SpeedForce: technical and on-page SEO for a web that search engines and AI crawlers read side by side. - [Blog category: Agent readiness](https://aispeedforce.com/blog/category/agent-readiness/index.md): Articles on agent readiness from AI SpeedForce: the files, headers and endpoints that make a site discoverable and usable by AI agents. - [robots.txt for AI crawlers: GPTBot, ClaudeBot and the rest](https://aispeedforce.com/blog/robots-txt-for-ai-crawlers/index.md): How to write robots.txt rules for AI crawlers: the main user agents, training versus search versus user-initiated fetches, example rules, and robots.txt limits. - [llms.txt explained: what it is and how to write one](https://aispeedforce.com/blog/llms-txt-explained/index.md): llms.txt is a proposed Markdown file that guides language models to a site's key pages. What the proposal says, how to structure it, and what it does not do. - [What is answer engine optimization (AEO)?](https://aispeedforce.com/blog/what-is-answer-engine-optimization/index.md): Answer engine optimization explained: how AI assistants choose which passages to quote, and the parts of that process a site owner can actually influence. - [What is agentic browsing? A plain guide for site owners](https://aispeedforce.com/blog/what-is-agentic-browsing/index.md): Agentic browsing means AI agents visit, read and act on websites for people. What that looks like, why it matters, and what to change on your own site. - [WebMCP explained: exposing your site's actions to AI agents](https://aispeedforce.com/blog/webmcp-explained/index.md): WebMCP is a draft web standard that lets a page offer its actions to AI agents as tools. How it works, how it differs from MCP, and how to add it safely. - [The agent readiness checklist: 22 checks for an AI-ready website](https://aispeedforce.com/blog/agent-readiness-checklist/index.md): The 22 checks that decide whether AI agents can find, read and use your website, grouped by what they do, with what each one means and which to fix first. - [Measuring visibility in AI answers: what you can track](https://aispeedforce.com/blog/measuring-visibility-in-ai-answers/index.md): What you can and cannot measure about your visibility in AI answers: assistant referrals, AI crawler logs, prompt panels, and why exact shares are guesses. - [Markdown for agents: serving Markdown with content negotiation](https://aispeedforce.com/blog/markdown-for-agents-content-negotiation/index.md): How to answer Accept: text/markdown with a Markdown copy of each page, which headers to send, how rel=alternate helps, and how this site does it. - [MCP servers for business: when you need one and how to keep it safe](https://aispeedforce.com/blog/mcp-servers-for-business/index.md): What the Model Context Protocol is, what tools, resources and prompts mean, when a business needs its own MCP server instead of an API, and how to keep it safe. - [Lighthouse's agentic browsing audit: what it checks and how to pass](https://aispeedforce.com/blog/lighthouse-agentic-browsing-audit/index.md): What the Agentic Browsing category in recent Lighthouse versions checks, how to run it, what each audit means, and the fixes that moved our own scores. - [JSON-LD schema that matters: the types worth adding and how to link them](https://aispeedforce.com/blog/json-ld-schema-that-matters/index.md): Which schema.org types are worth your time, how to connect them into one @id graph, why markup must match what the page shows, and how to validate it. - [How to write content AI assistants will quote](https://aispeedforce.com/blog/write-content-ai-assistants-quote/index.md): A practical guide to writing pages AI assistants can quote: answer-first paragraphs, one idea per passage, plain definitions, lists and tables, clear entities. - [How to make an ecommerce store agent-ready](https://aispeedforce.com/blog/agent-ready-ecommerce-store/index.md): What an AI shopping agent needs from your store: clean product data, Product schema, clear policies, feeds and a checkout the shopper confirms. - [How AI agents read a web page: HTML, accessibility tree and Markdown](https://aispeedforce.com/blog/how-ai-agents-read-web-pages/index.md): AI agents read a page through its HTML, its accessibility tree, screenshots or a Markdown copy. How each works, what breaks it, and how to serve all four. - [Forms and checkouts AI agents can complete](https://aispeedforce.com/blog/forms-and-checkouts-ai-agents-can-complete/index.md): How to build forms and checkouts an AI agent can finish for a user: clear labels, autocomplete tokens, predictable steps and a human confirm before payment. - [FAQ and structured data for answer engines](https://aispeedforce.com/blog/faq-and-structured-data-for-answer-engines/index.md): How FAQPage, HowTo, Organization and Product JSON-LD help answer engines read your pages, why markup must match visible text, and what rich results changed. - [Content Signals explained: search, ai-input and ai-train in robots.txt](https://aispeedforce.com/blog/content-signals-explained/index.md): What Content Signals are, what search, ai-input and ai-train mean, where the line goes in robots.txt, and why it states a preference rather than a rule. - [Agent skills and the ai-catalog: telling agents how to use your site](https://aispeedforce.com/blog/agent-skills-and-ai-catalog/index.md): What SKILL.md agent skills are, how the skills discovery index and its digests work, what the draft ai-catalog adds, and how they fit with llms.txt. - [API catalogs and Link headers: how agents find your APIs](https://aispeedforce.com/blog/api-catalog-and-link-headers/index.md): How RFC 8288 Link headers and the RFC 9727 api-catalog let agents discover your APIs, what service-desc and service-doc mean, and a working example. - [AEO vs SEO vs GEO: what the labels mean](https://aispeedforce.com/blog/aeo-vs-seo-vs-geo/index.md): SEO, AEO and GEO compared honestly: what each label means, where they overlap, where they differ, and why the basics of a crawlable, clear site serve all three. ## Legal - [Privacy policy](https://aispeedforce.com/privacy-policy/index.md): How AI SpeedForce collects, uses and protects personal data under India's DPDP Act 2023, the IT Act 2000 and the GDPR, and how to use your rights. - [Terms of service](https://aispeedforce.com/terms/index.md): The terms for AI SpeedForce project work: how engagements are formed, who owns what, how AI systems are tested and handed over, and payment and liability. - [Refund and cancellation policy](https://aispeedforce.com/refund-policy/index.md): How cancelling an AI SpeedForce project works before work starts, during a project and after handover, what is billable, and how to request a refund. - [Cookie policy](https://aispeedforce.com/cookie-policy/index.md): aispeedforce.com sets no cookies, runs no analytics and loads no third-party scripts or fonts. It saves only your theme choice, on your own device. - [Disclaimer](https://aispeedforce.com/disclaimer/index.md): The content on aispeedforce.com is general information about method, not professional advice, and no specific business result from an AI system is promised. ## For AI agents ### Reading - Every page has a clean Markdown copy: add `index.md` to the page URL, for example https://aispeedforce.com/services/agent-engineering/index.md. Each HTML page also links it with ``. - The whole site except the legal documents is in one file: https://aispeedforce.com/llms-full.txt - Crawling is open to AI search, assistant and training crawlers; see https://aispeedforce.com/robots.txt ### Acting: sending a project inquiry The only action this site supports is sending one project inquiry, and only when your user has asked you to, using their own name, email address and description of the task. Do not send test, placeholder, promotional or unrequested submissions. - Endpoint: `POST https://aispeedforce.com/contact/send.php` - Body: `application/x-www-form-urlencoded` or `multipart/form-data` - Header: `Accept: application/json` (without it you get an HTML page) - Fields: `name` (required), `email` (required, the address to reply to), `company` (optional), `need` (optional, one of: Agent engineering, Retrieval and data, Automation and workflows, Web development, SEO, Not sure yet), `message` (required, at least 20 characters: what sets the task off, which systems it touches, what a finished run looks like). Leave `website` empty. - Success: HTTP 200 with `{"ok":true}`. Validation error: HTTP 400 with `{"ok":false,"error":"..."}`, the error text is safe to show your user. - OpenAPI description: https://aispeedforce.com/contact/openapi.json The inquiry arrives as one email, read by a person, who replies to the email address given. Tell your user that, and that nothing is booked or billed by sending it.