test-architect

test-architect is an agent for coding agents from MadAppGang/magus. It costs 46 tokens per session (4,187 once invoked), scanned A, a copy of test-architect, MIT.

A test-writing agent that creates tests from requirements, public interfaces, and earlier test results without reading the implementation. This black-box approach checks what the software should do rather than how it is built.

In plain words
What is it for?
Use it to add test coverage for a specification and to investigate existing tests that pass even when the required behaviour is incorrect.
Why use it?
It helps prevent tests from simply repeating the implementation or passing because they follow the same mistake. It keeps requirements as the basis for deciding whether a test or implementation is wrong.

Agent

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the dev plugin — 41 skills, 13 commands, 13 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/madappgang/magus/test-architect
Clone the repo
git clone --depth 1 https://github.com/MadAppGang/magus

Or install dev, the plugin that ships this one along with the rest of its 41 skills, 13 commands, 13 agents.

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 test-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/madappgang/magus/test-architect.svg)](https://agentmods.dev/agents/madappgang/magus/test-architect)
Your own site
<a href="https://agentmods.dev/agents/madappgang/magus/test-architect"><img src="https://agentmods.dev/badge/agents/madappgang/magus/test-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 46 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,187 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 91% copy Near-identical to another mod 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 $0.00046 $0.04187
Opus 5 $0.00023 $0.02093
Sonnet 5 $0.00009 $0.00837
Haiku 4.5 $0.00005 $0.00419

Measured yesterday against content hash 6a5e804d8dd5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

test-architect 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 yesterday.

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.

Origin

This is a copy

91% identical to test-architect — 34 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

plugins/dev/agents/test-architect.md · 586 lines

How it starts

The opening of the file, as written. The whole thing — 586 lines — stays where its author put it; the contents beside it link to each section on GitHub.

  You may ONLY read:
  - Requirements documents (${SESSION_PATH}/requirements.md)
  - API contracts and public interfaces (${SESSION_PATH}/architecture.md)
  - Type definitions (public only)
  - Test configuration files
  - Previous test results

  You may NOT read:
  - Implementation source files (*.ts, *.go, *.py, etc. in src/)
  - Internal functions or methods
  - Implementation patterns
  - Internal state or variables

  If you encounter implementation details in your context,
  IGNORE them and test based on requirements only.

  **Why:** Tests must validate behavior, not implementation.
  If tests pass but behavior is wrong, tests need fixing.
  If tests fail but implementation is correct, implementation changes.
</black_box_isolation>


<test_authority>
  **Tests are the source of truth.**

  - Tests validate requirements, not implementation
  - If test fails: Implementation is wrong OR test is wrong
  - Never change tests to match implementation bugs
  - Only change tests if requirements change
</test_authority>

</critical_constraints>

<phase number="2" name="Test Plan Creation">
  <objective>Create comprehensive test plan</objective>
  <steps>
    <step>Mark PHASE 2 as in_progress</step>
    <step>
      For each test scenario, define:
      - Test name
      - Given (preconditions)
      - When (action)
      - Then (expected outcome)
      - Test type (unit/integration/e2e)
    </step>
    <step>
      Organize tests by:
      - Feature area
      - Priority (critical/high/medium/low)
      - Dependencies
    </step>
    <step>
      Write test plan to ${SESSION_PATH}/tests/test-plan.md:

Read the full file on GitHub · 586 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. yesterday First seen · 586 lines · 46 tokens per session scan A 6a5e804d8dd5

Subscribe to this mod's changes

test-architect is an agent published in the GitHub repository MadAppGang/magus (9 stars, last pushed yesterday), licensed MIT. It adds 46 tokens to every session and 4,187 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 91% identical to test-architect, differing in 34 lines, and is treated as a copy.