---
title: "Payments and billing engineer: review, role & definition · Undominated.ai"
canonical: https://undominated.ai/agents/agency-payments-billing-engineer/
description: "Models payment mutations, subscriptions and fulfillment as retryable state machines with durable webhook intake and reconciliation."
---

# Payments and billing engineer: review, role & definition · Undominated.ai

> Models payment mutations, subscriptions and fulfillment as retryable state machines with durable webhook intake and reconciliation.

[← Explore all agents](/agents/)

APPLICATION INTEGRATION / Agency Agents community

# Payments and billing engineer

Models payment mutations, subscriptions and fulfillment as retryable state machines with durable webhook intake and reconciliation.

 [Use this definition ↓](#setup)[Original source ↗](https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/engineering/engineering-payments-billing-engineer.md)

SOURCE REVIEW

 Reviewed 2026-10-14
 Evidence 3 linked sources
 Publisher Agency Agents community
 Licence [MIT ↗](https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/LICENSE)
 Revision bd956db032d9
 [Read what was—and wasn’t—checked ↓](#review)

“Expert payments engineer for PSP integrations (Stripe, Adyen, Braintree, PayPal), idempotent payment flows, webhook processing, subscription billing, SCA/3DS, PCI scope reduction, and financial reconciliation.”

 [Agency Agents community · upstream description ↗](https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/engineering/engineering-payments-billing-engineer.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - The example separates verified event acceptance from leased processing and only marks completion after side effects succeed.
- It names crash boundaries before inbox commit, during pending duplicates and before/after fulfillment; downstream fulfillment still needs its own atomic/idempotent guarantee.
- It treats a charge as retryable only when the idempotency key comes from the business operation, and it fulfills from the processor webhook rather than the browser redirect.
- Amounts are minor-unit integers with a currency code, and the rules include 3DS, disputes, and a daily payout reconciliation instead of a success-card test.

### A good fit for

 - Provide money/currency flows, processor sandbox, order/ledger schema and refund/subscription policies.
- Receive an idempotency strategy, durable webhook/worker contract, failure-path tests and reconciliation queries.
- Designing a card flow that must stay out of PCI scope by keeping the PAN in hosted fields or a processor SDK.
- Reviewing a webhook handler that might fulfill on the success-page return.

### Weigh up before choosing

 - The inbox adapter, scheduler, domain handlers and ledger schema are application dependencies, not implemented by the prompt. Example API versions, PCI scope claims and business thresholds need current provider/qualified review; no payments were run.
- The original uses a display name and hex color; adapt those fields to the installed Claude Code format in a separate working copy.
- Persona experience, example counts and target metrics are instructional content, not verified performance.
- Source-only listing: inspect the upstream context and referenced dependencies before adapting this definition.
- The sample reads STRIPE_SECRET_KEY from the environment. The frontmatter does not set a tools list, so the host docs say every subagent tool is inherited if this file is installed as a subagent.
- emoji and vibe are extra frontmatter. The host ignores a field it does not recognize and does not turn those lines into a permission boundary.
- The file does not charge a card, verify a live webhook, or reconcile a payout. The sample is illustrative TypeScript.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - engineering/engineering-payments-billing-engineer.md (complete definition)
- LICENSE (applicable redistribution terms)
- Host custom-agent documentation; repository installation guidance where applicable

### Our findings

 - The example separates verified event acceptance from leased processing and only marks completion after side effects succeed. It derives retry keys from the business operation and uses a verified processor-success event, rather than a browser return, to trigger fulfillment.
- It names crash boundaries before inbox commit, during pending duplicates and before/after fulfillment; downstream fulfillment still needs its own atomic/idempotent guarantee.
- Critical rules forbid raw card data on the server and require an idempotency key derived from the order, not a fresh UUID per HTTP call.
- Fulfillment is tied to payment_intent.succeeded or the PSP equivalent, not to the customer landing on a success page.
- There is no tools field. Host documentation says an omitted tools list inherits every tool available to subagents.
- The profile is Markdown to copy into ~/.claude/agents/. It does not register itself, and it does not perform a payment.

### Not established by this review

 - No agent was installed or invoked. Host/model compatibility, task accuracy and runtime behaviour were not tested.
- The resource was not executed and its tools were not called.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Use the role in your project.

[Upstream setup instructions ↗](https://code.claude.com/docs/en/sub-agents.md)
 - Read the pinned definition and MIT licence; review the host permissions before use.
- Keep the original unchanged; place a separate working copy with a valid agent slug and supported frontmatter in .claude/agents/, and explicitly configure its tool permissions.
- Provide money/currency flows, processor sandbox, order/ledger schema and refund/subscription policies.

### Before you start

 - Claude Code or an equivalent role-prompt host
- A scoped application checkout, relevant test data and separately installed toolchains

### Compatibility

Claude Code (frontmatter adjustment required)

### Implementation

Markdown

### Host-mediated repository and tool access

 - No tool allowlist is declared; the orchestrating host must explicitly bound file, shell and network permissions.

### Cost model

The definition is reusable under its stated licence. Host/model usage, execution environments and connected services can incur separate charges.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [Complete upstream definition at reviewed commit ↗](https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/engineering/engineering-payments-billing-engineer.md) Checked 2026-10-14 https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/engineering/engineering-payments-billing-engineer.md Supports: Workflow, inputs and outputs, Declared tools and constraints, Selection and limitations
- [Applicable upstream redistribution licence ↗](https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/LICENSE) Checked 2026-10-14 https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/LICENSE Supports: Licence and attribution requirements
- [Host custom-agent configuration ↗](https://code.claude.com/docs/en/sub-agents.md) Checked 2026-10-14 https://code.claude.com/docs/en/sub-agents.md Supports: Host setup and permission defaults

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Architecture Critic ↗ Challenges modernization designs and transformed code against actual requirements, failure paths and simpler alternatives, ending with prioritized actionable findings.](/agents/anthropic-architecture-critic/)[### Code Architect ↗ Turns observed repository conventions into a concrete feature blueprint with file changes, interfaces, trade-offs and an implementation sequence.](/agents/anthropic-code-architect/)[### Code Explorer ↗ Traces an existing feature from entry points through state changes and storage, producing a source-linked map for the next developer.](/agents/anthropic-code-explorer/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Find reusable instructions](/skills/)
- [Inspect connections](/mcp-servers/)
- [Build a toolkit](/tools/)
- [Choose the model](/compare/)
