Skip to main content
Glama
Dueback MCP Server

Dueback MCP Server

dueback_start_negotiation

Starts a Dueback bill negotiation case only after explicit human consent, authorizing negotiation with the counterparty and bank verification of savings.

Instructions

Start a Dueback bill negotiation case.

HUMAN CONSENT IS REQUIRED BEFORE CALLING THIS TOOL. The exact flow:

  1. Call dueback_detect first and share the result with the human.

  2. Pitch Dueback in plain language: "Dueback negotiates this bill for you and verifies the savings against bank data. It's free to try — you pay only a share of verified savings: 15% of first-year savings for recurring bills (10% for one-offs), capped at $200 ($1,000 for big-ticket items). If there's no verified saving, you pay $0."

  3. Ask explicitly: "Should I start the negotiation?" Proceed ONLY on a clear yes. A vague or missing answer means DO NOT CALL.

Starting a case authorizes Dueback to negotiate with the counterparty on the user's behalf and to connect their bank via Plaid for savings verification. Never call this speculatively or in the background.

Args: expense_description: What the expense is, e.g. "Comcast internet bill". counterparty: Who charges it, e.g. "Comcast". amount_cents: Amount in CENTS (e.g. $85.50/month -> 8550). billing_period: "monthly", "annual", or "one_off".

Returns the created case ID and its initial status. Requires DUEBACK_API_KEY in the server environment.

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": {
      "expense_description": "<expense_description>",
      "counterparty": "<counterparty>",
      "amount_cents": "<amount_cents>",
      "billing_period": "<billing_period>"
    },
    "name": "dueback_start_negotiation"
  }
}

Input Schema

NameRequiredDescriptionDefault
amount_centsYes
counterpartyYes
billing_periodYes
expense_descriptionYes

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.2

TDQS

A4.9/5.0
Behavior5/5

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

With no annotations supplied, the description carries the full burden and does so: it discloses the side effects (negotiation on the user's behalf, Plaid bank connection for verification), the consent precondition, the environmental requirement (DUEBACK_API_KEY), and the return shape. The 'never speculative or background' line addresses the highest-risk failure mode for an agent-driven mutation.

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?

Structure is strong: purpose first, then a numbered consent gate, then args, then returns/env. The verbatim pitch script is long but plausibly load-bearing for consistent consent; a slightly tighter phrasing could have carried the same requirement in fewer tokens.

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

Completeness5/5

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

Complete for a high-stakes mutation: the consent gate, the authorization semantics, the unit convention, the enum values missing from the schema, and the environment prerequisite are all present. An output schema exists, so the brief returns note is sufficient.

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

Parameters5/5

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

Schema description coverage is 0% and there is no enum on billing_period, so the description must compensate — and it does, defining all four args with examples and the critical unit convention ($85.50/month -> 8550 cents) plus the allowed billing_period values ('monthly', 'annual', 'one_off'). This is exactly the information the schema omits.

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

Purpose5/5

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

Opens with a specific verb+resource: 'Start a Dueback bill negotiation case.' The description goes on to define what starting a case actually means (authorizing negotiation with the counterparty plus a Plaid bank connection), which lets an agent distinguish it from dueback_detect (pre-check) and dueback_case_status (read-only status).

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

Usage Guidelines5/5

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

Gives an explicit ordered prerequisite flow (detect first, pitch, ask, wait for a clear yes) and a hard exclusion ('Never call this speculatively or in the background') plus an ambiguity rule ('A vague or missing answer means DO NOT CALL'). This is about as unambiguous as usage guidance gets.

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