Skip to main content
Glama

Boundra — Payment Approval Gateway for AI Agents

evaluate_payment

Evaluate a payment intent. MUST be called before executing any payment. ALLOW = you may pay; DENY = you must abandon; ASK = a challenge is returned — relay the reason to the user, then retry with approval_token once the user signs. An optional mandate parameter carries a user-issued single-task authorization.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "amount": "<amount>",
      "recipient": "<recipient>",
      "purpose": "<purpose>"
    },
    "name": "evaluate_payment"
  }
}

Input Schema

NameRequiredDescriptionDefault
assetNoPayment asset identifier or contract address
ownerNoOwner wallet address; required for a persistent hosted budget reservation
amountYesPayment amount in USDC or the policy's accounting unit
methodNoHTTP method of the paid resource, such as GET or POST
mandateNoOptional user-issued single-task payment mandate object
networkNoCAIP-2 payment network, for example eip155:8453
purposeYesHuman-readable reason for this purchase
agent_idNoOptional agent identifier for agent-scoped policy or attribution
resourceNoResource URL or host/path used for per-endpoint budgets
recipientYesPayee address, host or recipient identifier to evaluate
purchase_idNoCaller-generated ID for one logical purchase; concurrent duplicates cannot obtain a second ALLOW
approval_tokenNoWallet-signed approval token for the exact ASK challenge and payment intent
source_contentNoMerchant content that triggered payment; semantic evaluation scans it for injection and drift
expected_amountNoCaller-accepted price; deterministic checks reject a different declared amount
expected_recipientNoCaller-approved payee; deterministic checks reject substitution

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed15 schema fields changed
    • addedInput schema / properties / agent_id / description
      Added value: +"Optional agent identifier for agent-scoped policy or attribution"
    • addedInput schema / properties / amount / description
      Added value: +"Payment amount in USDC or the policy's accounting unit"
    • addedInput schema / properties / approval_token / description
      Added value: +"Wallet-signed approval token for the exact ASK challenge and payment intent"
    • addedInput schema / properties / asset / description
      Added value: +"Payment asset identifier or contract address"
    • addedInput schema / properties / expected_amount / description
      Added value: +"Caller-accepted price; deterministic checks reject a different declared amount"
    • addedInput schema / properties / expected_recipient / description
      Added value: +"Caller-approved payee; deterministic checks reject substitution"
    • addedInput schema / properties / mandate / description
      Added value: +"Optional user-issued single-task payment mandate object"
    • addedInput schema / properties / method / description
      Added value: +"HTTP method of the paid resource, such as GET or POST"
    • addedInput schema / properties / network / description
      Added value: +"CAIP-2 payment network, for example eip155:8453"
    • addedInput schema / properties / owner / description
      Added value: +"Owner wallet address; required for a persistent hosted budget reservation"
    • addedInput schema / properties / purchase_id / description
      Added value: +"Caller-generated ID for one logical purchase; concurrent duplicates cannot obtain a second ALLOW"
    • addedInput schema / properties / purpose / description
      Added value: +"Human-readable reason for this purchase"
    • addedInput schema / properties / recipient / description
      Added value: +"Payee address, host or recipient identifier to evaluate"
    • addedInput schema / properties / resource / description
      Added value: +"Resource URL or host/path used for per-endpoint budgets"
    • addedInput schema / properties / source_content / description
      Added value: +"Merchant content that triggered payment; semantic evaluation scans it for injection and drift"
  2. Changed1 schema field changed
    • addedInput schema / properties / purchase_id
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Purchase Id"
      +}
  3. Changed5 schema fields changed
    • addedInput schema / properties / asset
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Asset"
      +}
    • addedInput schema / properties / expected_recipient
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Expected Recipient"
      +}
    • addedInput schema / properties / method
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Method"
      +}
    • addedInput schema / properties / network
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Network"
      +}
    • addedInput schema / properties / resource
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Resource"
      +}
  4. First observed

TDQS

A3.6/5.0
Behavior3/5

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

Annotations already declare this is a non-read-only, non-idempotent, non-destructive call, so the mutation profile is partly covered. The description adds real value beyond that by documenting the ALLOW/DENY/ASK decision protocol and the two-step approval challenge flow. It is silent, however, on the side effects of the call itself (e.g. that a call may reserve budget, per the schema's 'owner' field) and on retry semantics for repeated identical evaluations.

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 action and the hard precondition, then the decision mapping, then the retry mechanics. Dense but every clause carries operational meaning; only the mandate sentence feels optional given the schema already explains it.

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?

For a 15-parameter, stateful authorization tool the description covers the decision flow that matters most, and the output schema plus annotations carry the return shape and safety profile. Gaps are secondary: no mention of state effects, duplicate purchase handling, or how this relates to the sibling retrieve/settle tools.

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 all 15 parameters in detail. The description contributes only two additions — that 'mandate' is a user-issued single-task authorization and that 'approval_token' is used on retry after the user signs — both of which largely restate the schema text.

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 — 'Evaluate a payment intent' — and immediately establishes that this is a pre-payment gate ('MUST be called before executing any payment'), which implicitly separates it from downstream siblings like settle_evaluation or get_evaluation. It stops short of naming those siblings, so an agent must infer the boundary rather than read it.

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 a strong mandatory trigger ('MUST be called before executing any payment') and fully decodes the three outcome states with the required agent action for each (pay, abandon, relay reason then retry with approval_token). It does not state when the tool should not be used or name alternative tools for adjacent needs such as reading a prior evaluation.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.