Frequently asked questions

Sign in with Google or GitHub. We create an isolated Sandbox workspace and catalog for you. No card required. Then paste a price book on Score or go straight to Onboard / your workspace.

A real price book or packaging sheet: product names, SKUs if you have them, list prices, currencies, billing periods, and any seat or region rules. Messy text is fine. Atlas structures offers and flags conflicts before you publish.

Use the readiness score in the workspace. It highlights missing prices, conflicting offers, weak packaging, and other blockers. Raise the score, then publish when the catalog is good enough for agents to trust.

Publishing versions your approved commercial truth and projects it onto the protocol rails your plan allows. Sandbox stays watermarked and MCP-only. Pro serves live catalogs on all rails without the sandbox watermark.

Yes. Most sellers extract, score, and publish first, then connect agent traffic later. Sandbox is built for that dry run.

No. PairRail is not merchant of record for your customers. Quotes stay indicative unless your org separately binds. Settlement and fulfillment stay with your stack.

Atlas is seller-side infrastructure for approved pricing, packaging, and policy in machine-readable form. Agents discover and read those offers on protocols such as UCP, MCP, A2A, ACP, and UAP, with governance so they only see what you authorized.

API-first and AI-native product teams, developer-tool vendors, and B2B SaaS sellers who need catalogs agents can discover and price without scraping PDFs or guessing list prices.

Your commercial truth is structured, versioned, readiness-scored, and served on protocol rails agents can consume without one-off integrations. Pro adds evidence-linked provenance and production rails.

No. Protocol responses and quotes are indicative unless your organization separately executes a binding agreement. Out-of-policy asks escalate to humans instead of inventing a number.

Fail-closed exception handling. When a request exceeds published mandate, Atlas queues it for human approval instead of guessing. Pro includes the deal-desk queue; Enterprise adds multi-approver workflows.

Sandbox is free for validation: one sandbox catalog, MCP only, watermarked responses, and a hard 1,000 queries/month cap. Pro publishes up to three live catalogs, unlocks all protocol rails, includes 25,000 queries/month, and keeps serving agents with soft overage instead of a hard cut.

Pro is $59/month, or $49/month when billed yearly ($588/year). Overage is $2 per 1,000 queries after the included volume. Currency and interval toggles live on the pricing page before checkout.

No. Sign in and start extracting and scoring without a card. You only pay when you upgrade to Pro.

Yes. Use the billing interval toggle before checkout. After you subscribe, change plans in the payment partner portal or email [email protected].

Checkout runs through our payment partner (Dodo Payments). PairRail does not store full card details. You can manage or cancel in the billing portal linked to your account, or email [email protected].

No. Enterprise covers custom volume, SLA, SSO, residency, and multi-workspace needs under contract. Use Contact sales on the pricing page or email [email protected].

Each priced protocol or gateway call against your live catalogs typically counts as one query. Sandbox traffic uses the Sandbox quota and does not consume Pro included volume.

After 25,000 queries in a month, overage is $2 per 1,000 queries. Pro keeps serving agents; usage continues and overage bills through our payment partner.

Sandbox hard-caps at 1,000 queries/month. Upgrade to Pro to continue with included volume plus overage, or wait until the next month resets.

Discovery surfaces can stay public so agents find you. Taking action (reading live offers, pricing, and quotes) requires seller-approved machine access. You decide which machines get credentials; rotate or revoke anytime without republishing the catalog.

Sandbox exposes MCP with watermarked responses. Pro and Enterprise unlock UCP, MCP, A2A, UAP, and ACP production endpoints without the sandbox watermark.

It marks responses as non-production. Agents can probe your catalog, but the surface is clearly sandbox and should not be treated as live commercial truth.

No. Publish commercial truth once in Atlas. Protocol projections come from that catalog, so you are not maintaining five parallel price books.

Yes. Homepage protocol chips and the Protocols page open live inspector views for the five commerce rails (/.well-known/ucp, /mcp, /a2a, /protocols/uap, /protocols/acp). The UCP product feed is part of UCP, not a separate rail.

Yes — as API documentation, not a sixth protocol rail. See /openapi.yaml for HTTP schemas used by the console and gateway. Commerce protocols remain UCP, MCP, A2A, UAP, and ACP.

Sign in with Google or GitHub. First login provisions an isolated org and a Sandbox catalog so you can extract, score, and publish without a credit card.

Paste or upload pricing content. Atlas structures products and offers, flags conflicts, and scores readiness. Sandbox includes a free Gemini share for exploration. Pro unlocks unlimited Gemini and multi-pass agentic extraction with evidence-linked provenance.

Sandbox is one seat. Pro supports a small team on the same seller workspace. Enterprise adds multi-workspace and SSO-style access under contract.

Yes. You approve which machines can act on your catalog. Pro includes scoped credentials with rotation and revocation, so you can cut access without republishing.

Sandbox keeps about 14 days of activity. Pro retains audit trail and catalog versions for 90 days. Enterprise can extend retention under contract.

Core services run on Google Cloud (Firebase Authentication and Hosting, Cloud Run, and Gemini / Vertex AI for extraction). See the Security page for the current posture summary.

You do. PairRail processes Customer Content only as needed to operate the service (hosting, protocol projection, extraction, backups). See the Terms of Use and Privacy Policy.

We do not sell Customer Content. Extraction uses configured model providers to transform content you submit for your workspace. Privacy explains how that processing works.

Email [email protected] with “Security disclosure” in the subject, or follow the Security page. For product misuse, see the Acceptable Use Policy.