--- title: "Questions about Tanso" canonical: https://tansohq.com/faq description: "What Tanso does, what is available now, what the CLI sends to Tanso, and where billing and MCP endpoints run." --- # Questions about Tanso ## What is available now? The agent-ready CLI, free. `npx @tansohq/agent-ready check yourproduct.com` reads your public pages, and `npx @tansohq/agent-ready test` sends a real agent through the task. Neither needs a Tanso account or key. See the [quickstart](https://tansohq.com/quickstart.md). Hosted: available now at app.tansohq.com (sign in): [sign in to the hosted dashboard](https://app.tansohq.com) for free checks and hosted test runs (5 free a month for a workspace a person owns, then $5.00 per completed run; a workspace an agent created that no person owns gets none). Agent Monetization is open-source billing, distributed as tanso-oss and run on your own infrastructure. The homepage demo uses authored sample data. ## Does the CLI send my data to Tanso? No. The CLI sends nothing to Tanso: no account, no key, no telemetry. It runs on your machine and sends requests only to your product and, for test, to Anthropic through your own Claude Code, and to AgentMail if you set `AGENTMAIL_API_KEY`. ## Does contacting Tanso start an audit? No. It starts a conversation, without creating an account, running an audit, or authorizing a purchase. ## Must an agent sign up or buy to succeed? No. The intended task defines success. An accurate evaluation can end without signup. Human approval can be part of a successful workflow. ## Where do billing and MCP endpoints run? On self-hosted instances, not this website. Tanso finds where your customers' AI agents get stuck, and fixes it. Tanso connects Web, Onboarding, and Monetization around the customer's intended task. ## Try it now npx @tansohq/agent-ready check yourproduct.com Free. No signup. The check reads your public pages the way an agent does, shows where it gets stuck, and writes one fix prompt per gap. [Give it to your coding agent](https://tansohq.com/quickstart.md). - CLI: available now, free - Hosted: available now at app.tansohq.com (sign in). A person [signs in at app.tansohq.com](https://app.tansohq.com) with an email code. Checks are free and unlimited. In a hosted test run, a real agent runs a saved task in the product on Tanso's model, and the outcome is verified separately. Hosted test runs: 5 free a month per workspace, then $5.00 per completed run. Inconclusive runs and runs that fail on Tanso's side are free. An agent can buy runs with POST /v1/purchases within a monthly cap ($50 by default, at most $200) that a person approves once through an approval link and Stripe Checkout. ## Products - **Agent Usability** ([quickstart](https://tansohq.com/quickstart.md)): the agent-ready CLI, @tansohq/agent-ready on npm. `check` reads public pages and `test` sends a real agent, your own Claude Code, through the task and checks the result. The CLI needs no Tanso account, key, or signup, and sends nothing to Tanso. Free, and open source under MIT ([source](https://github.com/tansohq/agent-ready-cli)). - **Agent Monetization** ([tanso-oss](https://github.com/tansohq/tanso-oss)): self-hosted plans, usage billing, credits, and access controls, run on the operator's own infrastructure. ## Availability The agent-ready CLI is available now and free. The CLI needs no Tanso account, key, or signup, and sends nothing to Tanso. A person [signs in at app.tansohq.com](https://app.tansohq.com) with an email code. Checks are free and unlimited. In a hosted test run, a real agent runs a saved task in the product on Tanso's model, and the outcome is verified separately. Hosted test runs: 5 free a month per workspace, then $5.00 per completed run. Inconclusive runs and runs that fail on Tanso's side are free. An agent can buy runs with POST /v1/purchases within a monthly cap ($50 by default, at most $200) that a person approves once through an approval link and Stripe Checkout. Free hosted runs need a workspace a person owns. A workspace an agent created through POST /v1/signup has none: its first hosted run answers 402 payment_required with an approval link, and the person who approves and pays takes ownership of the workspace. A person can also take ownership by signing in and entering the claim code on the Account page. See [Access and hosted runs](https://tansohq.com/auth.md) and [Pricing and availability](https://tansohq.com/pricing.md). The [interactive demo](https://tansohq.com/#demo) uses authored sample data. It is not a live evaluation of the visitor's product and not measured customer results. ## What Tanso is building - Web: product information, pricing, and documentation agents can find and accurately evaluate. An evaluation can succeed without signup. - Onboarding: setup, authorization, and integration paths leading to a useful result, with supported human approval where needed. - Monetization: plans, credits, usage, and authorized purchase handoffs connecting what agents use to what customers pay. ## The approach Understand the intended task and the observed problem. Distinguish a blocker from a harmless retry. Make a targeted change to the relevant information or workflow. Retest under comparable conditions and verify the actual result. An error is not automatically lost revenue. Not every task needs signup or payment. ## A real run, on tansohq.com The web evaluation was run on ten public SaaS sites (names withheld; public URLs, fetched the way an agent would): all ten publish llms.txt, none publish machine-readable pricing. Run on tansohq.com itself, four web probes shown: three passed and the pricing file warned because it declared no schema, revenue model or plans. Those were added and the same request re-run: all four pass. [Recording, 40 s](https://tansohq.com/video/real-run-web.mp4). Run the same checks on your product with the agent-ready CLI. For help with a fix, contact us. ## Existing open-source software tanso-oss provides self-hosted plans, usage billing, credits, and access controls. It supports the paid-usage part of the customer experience. Any self-hosted tanso-oss instance can expose agent signup, opt-in per account: POST /public/v1/catalog/{slug}/signup returns a customer, a free-plan subscription, and a customer-scoped API key in one call, and the instance serves a runbook at /agent-signup.md. The instance URL comes from the product an agent is buying from. Source: [tanso-oss on GitHub](https://github.com/tansohq/tanso-oss). - [Agent Usability overview](https://tansohq.com/product/agent-usability.md) - [Open-source overview](https://tansohq.com/product/agent-monetization.md) - [Source code](https://github.com/tansohq/tanso-oss) - [Documentation](https://tanso.mintlify.app) ## Contact Book a call at https://cal.com/katrina-laszlo/30-minute-meeting or email kat@tansohq.com. Contacting Tanso does not create an account or authorize a purchase. ## Public resources - [Quickstart](https://tansohq.com/quickstart.md): run the agent-ready CLI. - [AI Pricing Finder](https://tansohq.com/ai-pricing-finder.md): how AI products price usage, credits, commitments, and overages, with sources. This is not evidence of autonomous purchasing volumes. - [Finder methodology](https://tansohq.com/methodology.md) - [Current pricing data](https://tansohq.com/data/pricing-finder-current.json) - [Blog](https://tansohq.com/blog.md) - [Pricing and availability](https://tansohq.com/pricing.md) - [Questions](https://tansohq.com/faq.md) - [Access and hosted runs](https://tansohq.com/auth.md): what the CLI needs, and how agents sign up for and buy hosted test runs. - [Agent integration guide](https://tansohq.com/docs/agent-auth.md) - [Site information JSON](https://tansohq.com/api/agent) - [Full site context](https://tansohq.com/llms-full.txt) Contact: kat@tansohq.com.