MandateShield AI Payment Evidence

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

The description states 'This v1 entry check is non-executable and always returns enforcement_authorized=false', which implies a read-only stub. However, annotations mark readOnlyHint=false, suggesting potential side effects. This is a direct contradiction between the description and annotation, requiring a score of 1.

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

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

The description is three sentences, front-loaded with purpose, then behavioral disclosure, then a security warning. Every sentence contributes essential information with no redundancy.

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

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

Despite 19 parameters and nested objects, the description covers the key context: what the tool does, its non-executable stub behavior, the alternative for production, and a security warning. The output schema documents return values, so no further return detail is needed.

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

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

The description adds high-level context that parameters represent a proposed purchase and policy facts, but does not detail specific parameters. Schema descriptions cover 68% of parameters and explain many fields (e.g., currency codes, atomic amounts, identifiers), so the description provides only marginal added value.

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

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

The description opens with a specific verb and resource: 'Use before an AI agent buys, subscribes, transfers value, calls a metered API, or accesses a paid resource when the user wants a payment-authority check. Analyze whether the proposed purchase fits supplied policy facts.' This clearly states the tool's purpose and distinguishes it from the sibling 'strict cryptographic tool' for production gates.

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

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

Explicit when-to-use guidance is provided ('Use before an AI agent buys, subscribes...'), and an alternative is named ('use the strict cryptographic tool for a production gate'). Safety guidance ('Never send payment credentials or private keys') further clarifies usage constraints.

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

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

Annotations already declare readOnly, idempotent, and non-destructive behavior. The description adds critical disclosures beyond annotations: 'Evidence references are not independently verified,' 'projection_fields_valid is not full protocol conformance,' and 'always returns enforcement_authorized=false under assurance.' These set clear expectations for agents and are not repeated in annotations.

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

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

The description is four sentences long, front-loaded with the use case, and each sentence carries substantive constraints. It avoids fluff, but the dense jargon and compound phrasing make it less crisp than ideal.

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

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

Given the tool's complexity (4 parameters, nested objects, multiple protocol adapters), the description covers the trigger conditions, adapter-specific identity requirements, and the key limitation (no authority verification). It does not explain the output shape, but an output schema exists. This is sufficient for an agent to select and use the tool appropriately.

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

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

Schema description coverage is only 25%, so the description must compensate. It does clarify adapter selection through the enumerated protocols and explains payee_identity binding requirements (network+payTo for X402, service-origin+method for MPP). However, it does not explain most context fields (e.g., mandate_id, agent_id, asset_id, idempotency_key) or the optional selection parameter, leaving notable gaps.

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

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

The description opens with a specific trigger ('Use when an agent encounters...') and states the action: 'Map those documented fields into a deterministic MandateShield purchase envelope.' This clearly identifies the function as projection/normalization and differentiates it from sibling authority-checking tools by explicitly noting it 'never verifies delegated authority or a payment credential.'

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

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

The description explicitly tells when to use the tool: 'Use when an agent encounters an AP2 terminal closed-payment projection, x402 v2 PAYMENT-REQUIRED offer, or explicitly profiled MPP Payment challenge.' It also provides protocol-specific identity requirements and cautions that 'merchant_id alone is insufficient.' It does not name sibling tools directly, but the 'never verifies authority' phrase implies the boundary.

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

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

Annotations disclose side effects (not read-only, idempotent, not destructive), and the description adds important behavioral context: it fails closed for JWS, AP2-shaped SD-JWT, and TAP-shaped evidence, and creates only a short RESERVED authorization with cumulative-budget allocation. It further clarifies it never executes payment, aligning with openWorldHint=false and idempotentHint=true.

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

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

The description is three dense sentences, each carrying distinct information: usage constraints, validation behavior, and side-effect/processing boundary. There is no filler or repetition, and the front-loaded 'Use only for' immediately anchors the agent on the correct use case.

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

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

For a highly complex tool with deep nested schema and an output schema, the description adequately covers success behavior (qualifying ALLOW creates RESERVED authorization), failure behavior (fail closed for unsupported evidence), and the external processor flow. It is complete enough to position the tool within a broader payment pipeline without needing to explain return values.

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

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

The schema description coverage for the tool description is 0%, and the description does not directly explain the 'envelope' or 'evidence' parameters. It provides high-level prerequisites like registered mandate and exact final purchase, but these are not mapped to the actual parameters, leaving the agent to rely solely on the schema for field-level meaning.

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

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

The description clearly identifies the tool's purpose: a production pre-payment authority gate that verifies cryptographic payment evidence. It distinguishes itself from sibling tools by explicitly scoping out full AP2 checkout/delegate-chain and Visa TAP structured-field/trust-store processing as external, making its role as the fail-closed evidence verifier unambiguous.

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

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

The description begins with 'Use only for a production pre-payment authority gate after...' and lists precise preconditions (registered mandate, pinned issuer key, fresh challenge, VERIFY-scoped key, exact final purchase, supported signed evidence). It explicitly states what remains external and clarifies that the tool never executes payment, directing the agent to a separate trusted gateway for the processor step.

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