TWZRD Agent Intelligence

check_listing

Read-onlyIdempotent

Pre-spend check of one product against the published listing cards. Free, read-only.

Returns unknown_seller when no published card covers the product (not a clean seller),
expired when the card is past expires_at, check_failed when the card cannot be verified
or the declared price is above the advertised one, and advertised when a current card
matches. It never authorizes a payment: authorizes_spend is always false. needs_approval
is true for every result except advertised, and also for an advertised card that
advertises nothing, advertises no price, cannot check the declared price, or is
declared below its advertised price (an unverified discount, including zero). It does
not fetch the store page, open a checkout or move money.

Input Schema

Name	Required	Description
`market`	No	Optional subject market (e.g. US); must agree with the card when given.
`product`	No	Optional subject product; must agree with the card when given.
`variant`	No	Optional subject variant (e.g. Ink); must agree with the card when given.
`merchant`	No	Optional subject merchant; must agree with the card when given.
`product_url`	No	The store product URL the agent is about to buy from. Matched against the published card's next_hop URL, ignoring scheme, host case, query, fragment and a trailing slash. Another host is unknown_seller; a miss on the card's own site also returns covered_product_url.
`declared_unit_price`	No	Optional unit price the seller is asking, in USD, as a number or text, e.g. 78, 78.00, $78 or 78 USD; another currency is invalid. Above the card's advertised price is refused (phantom_markup_detected); below it is discount_unverified and needs approval.

Output Schema

Name	Required	Description
`claims`	No	The card's claims (name, value, state, reason). Unknown claims stay unknown.
`digest`	No	sha256 of the matching card's exact text (get_claim); null for unknown_seller.
`reason`	Yes	Why: e.g. no_card_for_product_url, expired_claim, phantom_markup_detected, current_card.
`result`	Yes	unknown_seller \| expired \| check_failed \| advertised.
`expires_at`	No	Earliest claim expiry on the matching card.
`artifact_id`	No	The matching card's artifact id; null when no card covers the candidate.
`price_check`	No	verified \| discount_unverified \| unverified \| absent \| above_advertised \| invalid, when a price was declared or checked.
`needs_approval`	Yes	True for every result except advertised, and for an advertised card that advertises nothing, advertises no price, cannot check the declared price, or is declared below its advertised price: a human approves before any spend.
`authorizes_spend`	Yes	Always false. advertised is current evidence, never permission to pay.

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedOct 20, 2026

addedInput schema / properties / declared_unit_price / anyOf

Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "number"
+  }
+]

changedInput schema / properties / declared_unit_price / description

Previous value: -"Optional unit price the seller is asking, in USD. Above the card's advertised price is refused (phantom_markup_detected); below it is discount_unverified and needs approval."New value: +"Optional unit price the seller is asking, in USD, as a number or text, e.g. 78, 78.00, $78 or 78 USD; another currency is invalid. Above the card's advertised price is refused (phantom_markup_detected); below it is discount_unverified and needs approval."

removedInput schema / properties / declared_unit_price / type
```
Removed value: -"string"
```

changedInput schema / properties / product_url / description

Previous value: -"The store product URL the agent is about to buy from. Matched exactly against the published card's next_hop URL, the same rule as the paid checkout brief."New value: +"The store product URL the agent is about to buy from. Matched against the published card's next_hop URL, ignoring scheme, host case, query, fragment and a trailing slash. Another host is unknown_seller; a miss on the card's own site also returns covered_product_url."

Changed2 schema fields changedSep 30, 2026

changedInput schema / properties / declared_unit_price / description

Previous value: -"Optional unit price the seller is asking, in USD. Above the card's advertised price is refused (phantom_markup_detected); below it is discount_unverified."New value: +"Optional unit price the seller is asking, in USD. Above the card's advertised price is refused (phantom_markup_detected); below it is discount_unverified and needs approval."

changedOutput schema / properties / needs_approval / description

Previous value: -"True for every result except advertised, and for an advertised card that advertises nothing, advertises no price, or cannot check the declared price: a human approves before any spend."New value: +"True for every result except advertised, and for an advertised card that advertises nothing, advertises no price, cannot check the declared price, or is declared below its advertised price: a human approves before any spend."

Changed1 schema field changedSep 29, 2026

changedOutput schema / properties / needs_approval / description

Previous value: -"True for every result except advertised, and for an advertised card that advertises nothing or cannot check the declared price: a human approves before any spend."New value: +"True for every result except advertised, and for an advertised card that advertises nothing, advertises no price, or cannot check the declared price: a human approves before any spend."

AddedSep 29, 2026

TDQS

A4.1/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint and idempotentHint, yet the description adds substantive behavior: it never authorizes a payment (authorizes_spend is always false) and spells out exactly when needs_approval is true, including edge cases like an advertised card with no price or a below-advertised declared price. These are decision rules an agent cannot derive from readOnlyHint alone.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the purpose and safety posture in the first sentence, then result semantics. The needs_approval sentence is long and nested ('advertises nothing, advertises no price, cannot check the declared price...'), which taxes readability, but nearly every clause carries decision-relevant information rather than filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Covers purpose, safety guarantees, and result/approval logic thoroughly for a 6-parameter check tool, and an output schema exists so return-value structure need not be restated. The remaining gap is routing among numerous plausibly overlapping siblings, which the description does not address.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents market/product/variant/merchant matching rules, product_url normalization, and declared_unit_price parsing. The description does not add parameter-level syntax or constraints beyond that, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: 'Pre-spend check of one product against the published listing cards,' plus scope qualifiers 'Free, read-only.' The purpose is unambiguous, but it never names or contrasts a sibling (e.g. get_shopping_check, get_merchant_card, get_publication), so an agent must infer the division of labor itself.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear context of use ('the store product URL the agent is about to buy from') and explicit boundaries ('does not fetch the store page, open a checkout or move money'). It stops short of naming alternative tools for adjacent jobs, so there is no explicit when-to-use-something-else guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All TWZRD Agent Intelligence tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.