testing-conventions

A testing guide for a Java repository with enforced rules for JUnit tests, build timeouts, network access, and architecture checks. TDD means writing tests to guide development, but this guide focuses on making tests pass the repository’s existing checks.

In plain words
What is it for?
Adding or fixing unit and integration tests, configuring fast isolated test fixtures, handling temporary files, and complying with Surefire and ArchUnit checks.
Why use it?
It prevents tests from failing because they run too long, access the network, use the wrong framework, or violate project structure rules.

Skill for Claude CodeCodex

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 skills/adamw7/tools/testing-conventions
Any agent
npx skills add adamw7/tools --skill testing-conventions
Clone the repo
git clone --depth 1 https://github.com/adamw7/tools

Made for: Claude Code, Codex.

Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,181 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00075 $0.01181
Opus 5 $0.00037 $0.00590
Sonnet 5 $0.00015 $0.00236
Haiku 4.5 $0.00007 $0.00118

Measured 2d ago against content hash 4c3ea9ccd9b8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

testing-conventions 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 2d 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/skills/testing-conventions/SKILL.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Testing Conventions Skill

Write unit and integration tests that satisfy the tools repo's enforced testing rules the first time. These rules are enforced by Surefire config, JUnit extensions, and ArchUnit architecture tests — a test that ignores them fails the build, not just review.

Hard rules (build fails otherwise)

Timeouts

  • 5 s per unit test. Surefire enforces a 5-second per-test timeout (root pom.xml). The bound is generous because the reactor builds in parallel (-T1C) and contending test forks stretch the cold-fork warmup; it is not a budget to spend. Keep unit tests fast — no real I/O, no sleeps, no heavy loops. A genuinely heavier test opts out with an explicit @Timeout and a comment explaining why.
  • Heavy shared setup (@BeforeAll etc.) has a looser 10-second limit (15 s under coverage). A fork that hangs outright is killed at 300 s (forkedProcessTimeoutInSeconds).

Unit tests run class-parallel

  • Surefire runs unit.test.parallelism (default 3) test classes at once per module; the methods within one class still run on a single thread, so a @BeforeEach-built fixture is never shared across threads.
  • A @TempDir is unique per class, so an ordinary test needs nothing extra.
  • Writing process-global state needs a guard, or the test corrupts whatever runs beside it:
    • a system property, or the deprecated PathValidator.setAllowedBaseDir → annotate the class @Isolated (org.junit.jupiter.api.parallel.Isolated); it then runs alone. ArchUnit fails the build if you forget. Confining a source with an AllowedPaths of its own writes nothing shared, and needs no annotation.
    • a fixture genuinely shared between classes (the embedded Derby database) → @ResourceLock("<name>"), which serialises just those classes. Put it on the shared base class; subclasses inherit it.
  • Debug a suspected ordering problem with -Dunit.test.parallelism=1.

Network is off for unit tests

  • The data module's NetworkOffExtension engages the Switch kill-switch before any test runs, so a unit test cannot open an outbound connection.
  • Anything needing the network is an integration test (*IT), gated behind the integration-tests profile and run by Failsafe — not Surefire.

Read the full file on GitHub · 94 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 94 lines · 75 tokens per session scan A 4c3ea9ccd9b8

Subscribe to this mod's changes

testing-conventions is a skill published in the GitHub repository adamw7/tools (11 stars, last pushed 4d ago), licensed MIT. It adds 75 tokens to every session and 1,181 once invoked, about $0.0004 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.