About agent-spend-policy-mcp
agent-spend-policy-mcp is a local MCP server that deterministically checks proposed agent spending actions against a supplied policy. It returns a decision and stable reason code without handling payments or making network requests.
Key features
- 01Evaluate actions against a supplied policy
- 02Return ELIGIBLE, DENY, or STEP_UP
- 03Provide stable reason codes
- 04Accept integer micro-unit money strings
- 05Perform no side effects or network requests
Use cases
- 01Dry-run proposed agent spending actions
- 02Check whether an action matches local policy
- 03Produce local evidence for a spending decision
MCP support
Exposes the `evaluate_spend_policy` tool over local stdio MCP. It returns policy evaluation outcomes only; payment authorization and execution are not supported.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
1Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| action | object | Yes | |
| action.rail | any | Yes | |
| action.asset | any | Yes | |
| action.agentId | any | Yes | |
| action.purpose | string | No | |
| action.expiresAt | string | Yes | |
| action.recipient | any | Yes | |
| action.authorityId | any | Yes | |
| action.payloadHash | any | Yes | |
| action.principalId | any | Yes | |
| action.amountMicros | any | Yes | |
| action.idempotencyKey | any | Yes | |
| policy | object | Yes | |
| policy.allowedRails | array | Yes | |
| policy.expectedAsset | any | Yes | |
| policy.maxDailyMicros | any | Yes | |
| policy.requirePurpose | boolean | No | |
| policy.stepUpAtMicros | any | No | |
| policy.expectedAgentId | any | Yes | |
| policy.allowedRecipients | array | Yes | |
| policy.maxPerActionMicros | string | Yes | |
| policy.expectedAuthorityId | any | Yes | |
| policy.expectedPayloadHash | any | Yes | |
| policy.expectedPrincipalId | string | Yes | |
| policy.expectedIdempotencyKey | any | Yes | |
| evaluationTime | string | Yes | |
| alreadySpentMicros | any | Yes |
Details
From the project READMEDingDawg Agent Spend Policy MCP
A small, local MCP server that deterministically evaluates whether a proposed agent spend action matches a supplied policy.
It returns one of ELIGIBLE, DENY, or STEP_UP, with a stable reason code.
ELIGIBLE is local policy evidence only. It is not payment authorization.
Safety boundary
This package does not hold funds, private keys, payment credentials, customer data, or settlement authority. It does not sign, send, settle, custody, or record payments. It makes no network requests.
A production payment adapter needs separate, independently verified controls for authenticated policy/action provenance, canonical payload hashing, durable atomic budget reservation and replay protection, customer-controlled signing, rail validation, and settlement reconciliation.
Install
npx -y @dingdawg/agent-spend-policy-mcp
Configure it as a local stdio MCP server:
{
"mcpServers": {
"dingdawg-agent-spend-policy": {
"command": "npx",
"args": ["-y", "@dingdawg/agent-spend-policy-mcp"]
}
}
}
Tool
evaluate_spend_policy accepts an evaluation time, a policy, a proposed action,
and the already-spent amount. All money is passed as integer micro-unit strings,
never JavaScript floating-point numbers.
The caller supplies the clock and already-spent value; therefore this tool is safe for dry runs and local evidence, not a replacement for a trusted payment or accounting system.
Agent contract
The versioned machine-readable contract is
capabilities.json. It describes the only tool this
package exposes, its required inputs, its three possible outcomes, and its
non-negotiable safety boundary.
- Transport: local stdio MCP
- Tool:
evaluate_spend_policy - Required inputs:
evaluationTime,policy,action, andalreadySpentMicros - Outputs:
ELIGIBLE,DENY, orSTEP_UP, each with a stable reason code - Side effects: none
- Credentials, payment execution, custody, signing, settlement, and network access: not supported
The manifest is package-source evidence for this release, not a promise of a
hosted agent-discovery endpoint. ELIGIBLE remains local policy evidence only,
not payment authorization.
Development
npm install
npm test
npm run pack:check
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.