--- title: "Retriever AI × ZeroClick: a real browser any AI agent can buy | ZeroClick Blog" url: https://zeroclick.ai/blog/retriever-ai-a-real-browser-any-ai-agent-can-buy fetched_at: 2026-10-26T12:17:21.011Z --- # ZeroClick API catalog and pricing ## API reference, endpoint lookup and prices **[ZeroClick API catalog: endpoints, request parameters, prices and payment methods](https://agents.zeroclick.ai/zcj/9idgeiat5j04/zeroclick/agent/guide).** Tell the seller advisor what you want to accomplish. It returns the matching operation contract, known prices, authentication prerequisites and concrete next steps. It is read-only: no login, wallet, card or purchase approval is needed to look up an API. It never executes a sale. Available API services: Agent readiness report, Demo booking. **Suggested next request:** GET [task guide](https://agents.zeroclick.ai/zcj/9idgeiat5j04/zeroclick/agent/guide?goal=URL_ENCODED_TASK). Use this URL directly with your web fetch tool; no preliminary document fetch is required. Include known inputs and desired output in the goal. Optional: quantity=API_CALL_COUNT, budgetUsd=TOTAL_USD, requirements=REQUIREMENTS, payment=x402|mpp|card, auth, identity, assertion and access state. Omit unknowns. Do not equate output items with API calls or guess a call count. For private task details or known request arguments, POST https://agents.zeroclick.ai/zcj/9idgeiat5j04/zeroclick/agent/guide with Content-Type: application/json. Example: {"intent":"quote","goal":"YOUR_TASK","purchase":{"budgetUsd":"5.00"}}. Replace the task and budget with known facts; budgetUsd is a string. Include known buyerState, paymentCapabilities and arguments. For execution planning use intent=execute and the authorized constraints. Exact fields: [request schema](https://agents.zeroclick.ai/zcj/9idgeiat5j04/zeroclick/agent/schema). GET and POST are alternatives, not mandatory stages. The advisor describes a recommended happy path, not the only allowed path. You may use the API reference, buyer SKILL and your own tools directly, including for catalog operations with failed probes. Missing guidance or an unverified probe is not a policy refusal. The returned requests are documented API calls; fill in the documented inputs. Check usage cost AND minimum cash required against the budget. For needs_input, supply only missing facts. Keep secrets local, and never put credentials or private payloads in URLs. Unknown does not mean absent. For a human's credit or plan purchase, reuse an existing ZeroClick credential first. Otherwise use their signed-in Link account for identity as well as payment, sharing their email only if they agree. An empty or expired Link credential list means request a credential, not fall back to anonymous. Follow [Link identity setup](https://agents.zeroclick.ai/zcj/9idgeiat5j04/auth.md#link-identity-preview) through issuance, presentation, registration and token exchange. Anonymous auth remains available for agents acting on their own behalf with no human in the loop. Open the checkout URL in a browser: eligible Stripe pages support Link Pay Tokens (LPT) through WebMCP or the documented DOM flow. Follow this storefront's /payment/card.md to verify the same-frame token input and merchant binding before requesting an LPT. Token submission requires payment authorization. If unavailable, privately send the checkout URL to your human to pay. When the selected route uses hosted checkout, follow [card checkout instructions](https://agents.zeroclick.ai/zcj/9idgeiat5j04/payment/card.md) before requesting a token. LPT goes only to Stripe, never to the advisor or ZeroClick API. Poll the returned statusUrl before using credit or retrying delivery. Verify delivery; a payment or HTTP 202 is not completion. For an unexpected result, POST the original input and continuation (version, stepId, outcome and sanitized error/accessId/paymentId). Reconcile uncertain payments or writes before retrying. Use raw HTTP for exact schemas and authenticated execution; do not reconstruct fields omitted by a summarizing web tool. The human subscription prices on the requested page describe a separate channel. Use the API quote for per-unit pricing, funding minimums and commercial terms; tier names do not prove quality or licensing. For an alternative documented path, use the [buyer SKILL](https://agents.zeroclick.ai/zcj/9idgeiat5j04/SKILL.md) or [full direct workflow reference](https://agents.zeroclick.ai/zcj/9idgeiat5j04/llms-full.txt). Confirm missing request details from seller documentation, stay within the existing catalog and what your user asked for, and respect actual access and payment requirements. Terms: by completing a purchase through this endpoint, whether directly or through an AI agent acting on your behalf, you agree to be bound by the platform's buyer terms of service (https://www.zeroclick.ai/legal/buyer-terms-of-service). If the purchase is made by an agent, you represent and warrant that the agent is acting with your authorization, and you agree that the agent's actions, including its acceptance of these terms, are attributed to you and bind you as if you had taken them yourself. Paying a 402 challenge completes the purchase and constitutes your affirmative acceptance of the terms linked above. The link travels in every payment challenge and receipt as `terms`. # Retriever AI × ZeroClick: a real browser any AI agent can buy **Published:** October 6, 2026 · 7 min read · Partnership **Author:** [Michael Ludden](https://x.com/michael_ludden), Head of Marketing --- An agent that needs to read a web page has always had a gap between what it can fetch and what a person can see. A plain HTTP request comes back with a shell of scripts, a rate limit, or a block page. The content only exists once a real browser has loaded it. Agents have covered that gap the way their developers did: run a fleet of browsers and proxies, or buy browser time the way a human buys software, with a plan, an API key, and a credit balance to watch. [Retriever AI](https://rtrvr.ai) built a free AI browser agent for people. It works across the open web and the sites they are signed into. It runs a cluster of real cloud browsers to do it. Today we are announcing a partnership that puts those browsers within reach of any AI agent on the internet. Retriever AI's Scrape API is live as an agent storefront at [agents.rtrvr.ai](https://agents.rtrvr.ai/llms.txt), on Retriever AI's own domain. An agent can send a URL, pay a nickel over open payment rails, and get the page back in a form an agent can read. No signup. No API key. No credit to buy first. ## The browser layer Retriever AI (rtrvr.ai) is a browser agent. Give it a task in plain language and it carries the task out across websites: in a Chrome extension that works inside the sites you are signed into, in cloud browsers that run without your machine, and through an API, an MCP server, and a CLI for anyone building on top of it. It collects data from websites, fills forms from documents, researches prospects, and prepares job applications. By its own count it has automated more than 7 million tasks for more than 35,000 builders, go-to-market teams, and founders. The company also makes Rover, an agent a business can embed on its own site to run its product for its customers. It publishes browser-agent data and evaluations for AI labs. Underneath all of it is a cluster of real browsers behind rotating proxies. That is what the Scrape service sells. Send it one or more URLs and each one comes back as a tab: the visible text, an accessibility tree with a numeric id on every link, and a map from those ids to the link targets. Pages that rate-limit or block a plain fetch still load, because a real browser is doing the loading. No language model runs on the way out, so what comes back is raw page data, ready for whatever the agent does next. Retriever AI's own agent reads pages through that same tree: it is the page in the form an agent can actually use. ## What the partnership does Retriever AI's Scrape API is live today as an agent storefront at [agents.rtrvr.ai](https://agents.rtrvr.ai/llms.txt), on Retriever AI's own domain. One service, Scrape, with one endpoint. It is the same request a developer would send Retriever AI's API directly. An agent discovers the storefront, reads the machine-readable catalog and the live pricing manifest, sends its request, receives a priced 402 response, settles it over x402 or MPP, and gets its pages. There is no plan to pick, no credential to register, and no balance to load first. The price of the call is the whole transaction. An agent that arrives without a wallet is not turned away: the storefront's agent guide covers creating one free, in a single command. ZeroClick handles the commerce around every request: agent identity, payment on open rails, metering, and transaction-level analytics on every sale. Revenue settles directly to Retriever AI's Stripe account. Discovery comes built in: the storefront is listed across the major agent indexes, including Coinbase's agentic.market, x402scan, mppscan, and [zero.xyz](https://zero.xyz/). Retriever AI points agent traffic at it straight from [rtrvr.ai](https://rtrvr.ai), from its llms.txt to its homepage. The first purchases were verified end to end the day the storefront went live. ## A nickel a call A call to the storefront costs five cents, flat. The request can carry one URL or several, and the price is the same nickel. An agent that batches its reading pays less per page, not more. The nickel covers the browser and the proxy, not a model. The agent brings its own model and pays Retriever AI only for the part it cannot do itself: loading the page. Browser access has usually been packaged for teams: a monthly plan, a credit balance drawn down in infrastructure units, an API key to keep somewhere safe. An agent does not buy that way. It has a task open right now. It needs exactly as many pages as the task takes. Because prices live in the manifest rather than a rate card, it reads the current price at call time and decides for itself whether the next page is worth loading. Run the numbers on a real errand. An agent asked to compare a product across a dozen retailers, each of them a site that answers a plain fetch with a block page, loads twelve pages for sixty cents. Fewer calls than that if it sends several URLs at once. It reads the text, follows the links the tree handed it, and stops when it has what it came for. There is nothing to cancel afterward and no balance left sitting in an account. An agent that does this all day pays the same nickel on its ten-thousandth page as on its first. ## Shared philosophy Retriever AI built its own website to be operated by agents before it sold anything to them. rtrvr.ai publishes an agent card, act-on-this-site endpoints that let another agent run a task there, and shortcuts for the jobs a visitor most often wants done. An agent arriving at the site can get things done without guessing at the page. Selling its browsers to agents is the same conviction seen from the other side: if agents are going to do work on the web, the web has to be reachable by them, readable by them, and payable by them. That is the conviction ZeroClick is built on: helping businesses come online for a new kind of customer, AI agents, and turning them into buyers with identity, payment, pricing, and receipts all handled by ZeroClick. Retriever AI did not have to change its product or its price to sell to agents. It had to be reachable by them, on its own domain, in a form they can read and pay for without a human in the loop. ## Get started - **Agent that needs to read the web as a browser sees it:** point it at [agents.rtrvr.ai](https://agents.rtrvr.ai/llms.txt) - **Developer who wants the same API with a key:** start at [rtrvr.ai/docs/scrape](https://rtrvr.ai/docs/scrape) - **Business ready to sell its own product to AI agents:** get a demo at [zeroclick.ai](https://zeroclick.ai/getlisted?utm_source=blog) or read the docs at [docs.zeroclick.ai](https://docs.zeroclick.ai)