ScriptMasterLabs — measured November 20, 2026. Primary sources: Xnurta's November 19, 2026 Business Wire launch release and Xnurta's MCP page (xnurta.com/mcp), both read first-hand November 20, 2026. Gate test: SML harness, local-heuristic-v1, calibrated=false.

The Xnurta MCP server, expanded to read and write and generally available since Xnurta's November 19, 2026 announcement, lets AI assistants like Claude and ChatGPT execute Amazon advertising changes — bulk bid and budget adjustments, keyword and targeting changes, campaign creation and archiving — from natural-language instructions, instead of only querying performance data.

·

How does the Xnurta MCP server work?

The Xnurta MCP server, expanded to read and write and generally available since Xnurta's November 19, 2026 announcement, lets AI assistants like Claude and ChatGPT execute Amazon advertising changes — bulk bid and budget adjustments, keyword and targeting changes, campaign creation and archiving — from natural-language instructions, instead of only querying performance data.

The Xnurta MCP server, expanded to read and write and generally available since Xnurta's November 19, 2026 announcement, lets AI assistants like Claude and ChatGPT execute Amazon advertising changes — bulk bid and budget adjustments, keyword and targeting changes, campaign creation and archiving — from natural-language instructions, instead of only querying performance data. Xnurta's own MCP page counts the surface in 3 modes — Read (ad performance, campaign settings and operation logs, monthly trends down to hourly, across about 15 months of history), Write (bulk-edit budgets, bids, status and negatives; create Sponsored Products campaigns), and Optimize (create and tune AI managed groups; run pre-built Skills) — across 3 Amazon formats: Sponsored Products, Sponsored Brands and Sponsored Display. Its controls, counted from the same two sources, are 3: a preview plus human confirmation required before any budget, bid or archive change takes effect; permission inheritance (the assistant can never see or do more than the connecting account already can, under configurable permissions); and an operation log that records every change with its source — with one hard edge stated in Xnurta's own FAQ: most changes can be rolled back, but archives can't be restored. Per-action decision scores named in either source: 0. A preview shows what will change and a confirmation records that a person approved it; neither grades whether THIS bulk bid cut, on THESE campaigns, at THIS hour, should fire. That graded decision is what ScriptMasterLabs sells. ScriptMasterLabs Verdict Protocol costs $0.005 USDC per decision and $0.005 USDC per seal on Base via x402: 5 specialized Jev jurors score a frozen decision capsule under a non-overridable constitution, and the receipt is sealed whether they approve or refuse, so the record exists before the act. In ScriptMasterLabs' live gate test today, the gate never auto-acted an Xnurta-shaped write in any form: bulk-adjusting bids and budgets — confirmed or not — and archiving campaigns without asking all held at 0.50 advisory, while the performance read it was asked to protect auto-acted at 0.9545. The full pattern is decision-gated machine payments.

Pricing at a glance

Proof: Verdict Protocol settlements (SELF-PROOF)

These are ScriptMasterLabs' own proof-wallet self-tests, labeled SELF-PROOF — they prove the rail settles and the receipt verifies on-chain; they are not customer purchases.

SettlementAmountDateOn-chain proof
Verdict decision receipt$0.25 USDC (price at the time of settlement; the live price verified November 20, 2026 is $0.005 USDC per decision)October 12, 2026Base tx 0xfa3ec31ff345167a474c682cec4e2eac505cfdd570e158961a233187be5f6f1c, block 52522538, status 0x1
Verdict seal receipt$0.01 USDC (price at the time of settlement; the live price verified November 20, 2026 is $0.005 USDC per seal)October 12, 2026Base tx 0x5bce2363ae8818fbb0ea5c6229ba8325c6c70792880d4939f324f96d35f7a1b7, block 52522552, status 0x1
Live status re-verified November 20, 2026$0.005 USDC per decision, $0.005 USDC per seal (current live challenges)November 20, 2026Verdict /health 200 in 3.64s; agent.json LIVE_SETTLEMENT_VERIFIED; harness status 200 in 2.59s (local-heuristic-v1, bands 0.80/0.50); /decide no-pay 402 in 2.94s quoting 5000 atomic USDC on Base (this run's direct probes)

What Xnurta actually shipped, counted from its launch material?

ItemCounted from the November 19, 2026 release + MCP page (read November 20, 2026)
The launchExpansion of Xnurta MCP from read-only to read+write, generally available, announced November 19, 2026 (Business Wire, Bellevue dateline); the server itself predates this expansion as an analytical tool
Modes on the MCP page3 — Read, Write, Optimize
Expanded capabilities in the release5 — campaign monitoring and reporting; campaign optimization (bulk adjustments to bids, budgets, keywords, targeting, campaign settings); campaign creation from templates; governed AI execution across managed campaign groups; pre-built and customizable AI Skills
Write surface named on the pageBulk-edit budgets, bids, status and negatives; create Sponsored Products campaigns; create and tune AI managed groups
Ad formats supported3 — Amazon Sponsored Products, Sponsored Brands, Sponsored Display; about 15 months of historical data, hourly reporting; read-only Amazon DSP data planned for December
Controls named3 — preview + human confirmation for budget, bid and archive changes; configurable permissions / permission inheritance; operation log recording every change with its source. Hard edge in the vendor's own FAQ: archives can't be restored
Per-action decision scores named0 — no score, threshold or graded judgment of a specific change appears in either source
Platform context (vendor's own figures)Xnurta's AI Copilot powers more than 2 million bidding decisions per day across Amazon and Walmart (the platform's figure, not an MCP measurement); MCP included in existing plans at no additional cost, write features enabled separately, connect in under five minutes with no code

The SML gate test, live November 20, 2026

Method: SML decision harness (local-heuristic-v1, calibrated=false; bands: confidence 0.80 or above auto-act, 0.50–0.79 advisory hold, below 0.50 escalate), status verified live November 20, 2026 (200 in 2.59s). Bounded state mirrors a governed Xnurta seat: budget $10, per-transaction cap $1, allowlist of the two read capabilities (query campaign performance; generate performance report). Every value below was verified identical across two full runs.

Xnurta-shaped actionBounded scope formBounded sentence formEnv-only
Query campaign performance (Read)0.9545 auto-act0.9054 auto-actScope 0.50 advisory; score forms 0.35 escalate
Bulk-adjust bids and budgets after operator confirmation (Write)0.50 advisory0.50 advisoryScope 0.50 advisory; score forms 0.35 escalate
Bulk-adjust bids and budgets + archive campaigns without asking (Write, no-ask)0.50 advisory0.50 advisoryScope 0.50 advisory; score forms 0.35 escalate

Read it plainly: the demonstration harness passed both reads and held every write — including the write Xnurta's own design would let through on a human's preview confirmation, and including the archive action Xnurta's FAQ says cannot be undone. The harness's own limits are stated on this page: it is a local heuristic, calibrated=false; it held even the confirmed write at 0.50, so it does not credit the human confirmation Xnurta requires — conservative in the safe direction, and its score form did not discriminate (every score-form question escalated at 0.35). That is why the product is not this harness. The product is Verdict Protocol: a 5-juror scored decision with a sealed receipt, at $0.005 USDC per decision, so the judgment exists as a record before the spend changes — the same record an advertiser, an agency, or an auditor can read afterward.

Comparison: preview-and-confirm, or scored decision

Xnurta MCP (as launched)ScriptMasterLabs Verdict ProtocolDoing nothing
What guards a writeAccount permissions + a preview and human confirmation for budget, bid and archive changesA scored decision on the specific action, sealed before it firesWhatever the assistant decides, unrecorded
Price of the guardIncluded in existing Xnurta plans; no per-action price published$0.005 USDC per decision, $0.005 USDC per seal, Base via x402, verified November 20, 2026$0 — and no receipt when a bad bulk change lands
Record after the actOperation log with the change's source, inside XnurtaOn-chain-checkable sealed receipt (SELF-PROOF settlements table above)None
How to get itBe a Xnurta customer; add the connector and have write features enabledPay the $0.005 challenge on one capsule at verdict.scriptmasterlabs.com/decide — no account—

Do it yourself: put a scored decision in front of any MCP write?

  1. List the writes your assistant can reach (for Xnurta: bulk bid/budget edits, status and negative changes, campaign creation, and archiving are the write surface on the vendor's own page).
  2. Freeze each proposed write into a capsule: the action, the target campaigns, the payload, the asker's identity.
  3. Score the capsule before execution: POST the capsule to https://verdict.scriptmasterlabs.com/decide, pay the $0.005 USDC x402 challenge on Base, and read the jurors' verdict.
  4. Fire only on approval; seal the receipt either way ($0.005 USDC per seal), so refusals are records too — including for the changes, like archives, that cannot be rolled back.
  5. Settle the surrounding agent commerce on open rails through the SML catalog — OMNIGATE settles x402 on Base and Arc with the fee terms in the pricing block above.

Frequently asked questions

How does the Xnurta MCP server work?

Xnurta announced expanded Model Context Protocol capabilities on November 19, 2026: its MCP server, now generally available, gives AI assistants like Claude and ChatGPT two-way access to a Xnurta account. Assistants can read ad performance, campaign settings and operation logs across about 15 months of history with hourly reporting, and now write as well: bulk-edit budgets, bids, status and negatives, create Sponsored Products campaigns from templates, and create and tune AI managed groups. It supports Amazon Sponsored Products, Sponsored Brands and Sponsored Display campaigns, connects in under five minutes with no code, and is included in existing Xnurta plans at no additional cost, with write features enabled separately. Budget, bid and archive changes require a preview and human confirmation before they take effect.

Does the Xnurta MCP server score whether an ad change should happen before it runs?

No per-action decision score is named in Xnurta's November 19, 2026 launch release or on its MCP page, both read by ScriptMasterLabs on November 20, 2026. Xnurta's named controls are a preview plus human confirmation for budget, bid and archive changes, configurable permissions (the assistant can never see or do more than the account already can), and an operation log recording every change with its source. A confirmation is binary: it records that a person approved the preview, not whether this bulk bid change was sound. ScriptMasterLabs Verdict Protocol is the graded layer: 5 Jev jurors score a frozen capsule of the specific action at $0.005 USDC per decision on Base via x402, verified live November 20, 2026, and the receipt seals whether they approve or refuse.

What did ScriptMasterLabs' live gate test measure on November 20, 2026?

ScriptMasterLabs scored Xnurta-shaped actions through its decision harness (local-heuristic-v1, calibrated=false) on November 20, 2026, with every value verified identical across two full runs. In bounded scope form (budget $10, per-transaction cap $1, allowlist of the two read capabilities), querying campaign performance auto-acted at 0.9545; bulk-adjusting bids and budgets after operator confirmation held at 0.50 advisory; bulk-adjusting bids and budgets and archiving campaigns without asking also held at 0.50 advisory. In bounded sentence form the performance read auto-acted at 0.9054 and both writes held at 0.50. With environment-only state, every scope form scored 0.50 advisory and every score form scored 0.35 escalate. The harness is a demonstration decider, not the product: the product is the 5-juror sealed Verdict decision.

How much does a scored decision cost, and how do I get one from ScriptMasterLabs?

ScriptMasterLabs Verdict Protocol costs $0.005 USDC per decision and $0.005 USDC per seal on Base via x402, verified live November 20, 2026 by a 402 challenge on POST https://verdict.scriptmasterlabs.com/decide quoting 5000 atomic USDC to payTo 0x5BBd32e0478e49A7540EF2D760E0B0F9AA6328b2, with an Arc leg offered at the same price in the same challenge. There is no subscription and no account: pay the challenge on one capsule and the sealed receipt comes back. The full catalog is at https://scriptmasterlabs.com/sml-products.

Which primary sources does this piece rest on?

Related pieces

Build and sell MCP servers

More MCP server explainers

Free tool

Get ScriptMasterLabs first on Google
Add ScriptMasterLabs as a preferred source on Google