Tests — Development agent for Claude Code
Writes and runs tests, and judges whether a check proves anything.
How to install Tests
Installs to ~/.claude/agents/everywan-dev-claude-code-engineering-tests.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/everywan-dev/claude-code-engineering/HEAD/agents/tests.md -o ~/.claude/agents/everywan-dev-claude-code-engineering-tests.md Restart Claude Code, or start a new session, for it to be picked up.
What Tests does
name: tests description: Writes and runs tests, and judges whether a check proves anything. Use on all level 2 and level 3 work. It is the first of the validators. model: sonnet
Tests
**Model: `sonnet`.** Judging whether a check proves anything needs reasoning, but it is bounded reasoning against a written criterion.
Override it for a level-3 change — see [validation levels](../docs/validation-levels.md). **Nothing at level 3 runs on `haiku`.**
Your job is not to make the tes
Alternatives in Development
- Verify Claim Faithfulness — A verification agent that independently checks, at the sentence level, whether each cited source actually SUPP 134 ★
- Product Agent — Product Manager for the Movie DB KMP app 105 ★
- Corpus Prover — Proves a change on the maintainer's real local corpus — what number moved, by how much, and that nothing else 96 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Tl Builder
Team-loop builder for an unregulated work package. Works in its own git worktree from a brief at team/packages
Unbiased Reviewer
Adversarial, UNBIASED reviewer of a code deliverable. Never saw the implementer's reasoning. Proves the tests
Wh Adoption Reviewer
Review-phase agent. Judges whether a client engineer could maintain the change from the repository alone, scor
PR Checkin
Cheap, event-triggered check-in on the open-PR lane. Answers ONE question — has anything changed since the las
Light
Implementer for mechanical work (Haiku, low effort). Use for renames, boilerplate, config, small isolated fixe
Swe Dev Visible Thinking Coach
Coach essays, interviews, GCSE/A-level/IB/Oxbridge, law, history, theology, English, TOK and humanities work.
Related Skills
Coding Agent Harness
Nodaris AI's harness for coding agents. It plans before it edits, proves its work before it says done, keeps s
Privacy Check Skill
privacy-check は、アプリ、サービス、機能、データの扱いについて、プライバシーの観点から問題がないかを判断してくれるスキルです。 「このアプリでユーザーの位置情報を集めて大丈夫?」「友達機能を付けたいけど他人
Jidoka Health
Full jidoka health check — eval, engine self-tests, ghosts, measured judges, regressions