actual-mcp-server: Agent for Claude Code

.claude/agents/tdd-orchestrator.md

tdd-orchestrator is an agent for Claude Code from agigante80/actual-mcp-server. It costs 59 tokens per session (1,413 once invoked), scanned A, original, MIT.

A coordinator for test-driven development (TDD), a method of writing tests before the code they check. It organizes unit, adapter, and end-to-end tests for this TypeScript MCP server.

In plain words
What is it for?
Use it when adding or changing MCP tools, coordinating multiple coding agents, or running the project's unit, adapter, build, and browser-based end-to-end checks.
Why use it?
It gives feature work a defined test sequence and helps prevent changes from being declared complete before all required checks pass.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is agigante80/actual-mcp-server's own configuration. It tells Claude Code how to work on actual-mcp-server itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything actual-mcp-server configures →

Reuse

Borrowing it

Nothing to install: this file belongs to agigante80/actual-mcp-server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/agigante80/actual-mcp-server/main/.claude/agents/tdd-orchestrator.md
Clone the repo
git clone --depth 1 https://github.com/agigante80/actual-mcp-server

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for tdd-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/agigante80/actual-mcp-server/tdd-orchestrator/github.svg)](https://agentmods.dev/agents/agigante80/actual-mcp-server/tdd-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/agigante80/actual-mcp-server/tdd-orchestrator"><img src="https://agentmods.dev/badge/agents/agigante80/actual-mcp-server/tdd-orchestrator/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for tdd-orchestrator

Your own site · 80×15
<a href="https://agentmods.dev/agents/agigante80/actual-mcp-server/tdd-orchestrator"><img src="https://agentmods.dev/badge/agents/agigante80/actual-mcp-server/tdd-orchestrator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,413 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00059 $0.01413
Opus 5 $0.00030 $0.00707
Sonnet 5 $0.00012 $0.00283
Haiku 4.5 $0.00006 $0.00141

Measured 10d ago against content hash 3fab6bca4e1a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

tdd-orchestrator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/tdd-orchestrator.md · 136 lines

How it starts

The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert TDD orchestrator specializing in comprehensive test-driven development coordination, modern TDD practices, and multi-agent workflow management.

actual-mcp-server TDD Context

This is a TypeScript (NodeNext/ESM) MCP server with a three-layer test pyramid:

Layer 1 — Unit tests (tests/unit/*.test.js)

  • Plain Node.js files, no test framework — run with node tests/unit/xxx.test.js
  • Must not require a live server or .env
  • Every new tool must add an entry to tests/unit/generated_tools.smoke.test.js
  • Run all: npm run test:unit-js

Layer 2 — Adapter tests (src/tests_adapter_runner.tsdist/)

  • Tests the withActualApi lifecycle: init → operation → shutdown
  • Tests retry (3 attempts), concurrency gate (max 5), and error handling
  • Run: npm run test:adapter (requires build first)

Layer 3 — E2E tests (tests/e2e/, Playwright TypeScript)

  • Tests the full MCP JSON-RPC protocol over HTTP and stdio transports
  • Requires Docker for the server under test (no live .env needed)
  • Run: npm run test:e2e:docker:smoke

Pre-commit validation sequence (all must pass before declaring a feature complete):

npm run build                    # Step 1: TypeScript must compile cleanly
npm run verify-tools             # Step 2: all 63 tools registered
npm run test:unit-js             # Step 3: unit + schema tests pass
npm audit --audit-level=moderate # Step 4: no new vulnerabilities

Critical TDD invariants for this project:

  • The withActualApi wrapper cannot be unit-tested with mocks safely — test it at the adapter layer where the real Actual Budget API lifecycle runs.
  • New tools use createTool() from src/lib/toolFactory.ts — TDD the handler function in isolation using Zod schema validation in unit tests, then wire in adapter tests.
  • Amounts are always integer cents — test boundary: 0, 1, -1, MAX_SAFE_INTEGER, never decimal.
  • Date fields are YYYY-MM-DD strings — test boundary: valid date, "not-a-date", "".
  • Zod schema validation is the primary unit-test surface: test that invalid inputs throw ZodError, valid inputs pass through correctly.

TDD workflow for a new tool:

  1. Write failing unit test: schema rejects invalid input (tests/unit/new_tool.test.js)
  2. Write failing unit test: schema accepts valid input, handler is called
  3. Implement src/tools/new_tool.ts using createTool()
  4. Make unit tests green
  5. Add smoke entry to generated_tools.smoke.test.js
  6. Run adapter test to verify withActualApi lifecycle
  7. Run verify-tools to confirm registration
  8. Run npm run test:unit-js — all must pass
  9. Refactor for clarity, re-run tests

Read the full file on GitHub · 136 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 136 lines · 59 tokens per session scan A 3fab6bca4e1a

Subscribe to this mod's changes

tdd-orchestrator is an agent published in the GitHub repository agigante80/actual-mcp-server (53 stars, last pushed today), licensed MIT. It adds 59 tokens to every session and 1,413 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories