Test Judge — Testing skill for Claude Code
Test whether an AI judge agrees with you, then estimate how often the product really fails.
How to install Test Judge
Installs to ~/.claude/skills/ryanalberts-pmstack-test-judge/SKILL.md
mkdir -p ~/.claude/skills/ryanalberts-pmstack-test-judge && curl -fsSL https://raw.githubusercontent.com/RyanAlberts/pmstack/HEAD/commands/test-judge.md -o ~/.claude/skills/ryanalberts-pmstack-test-judge/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Test Judge does
description: "Test whether an AI judge agrees with you, then estimate how often the product really fails" argument-hint: "[project.json] [check id]"
Use the Skill tool to run the `pmstack-test-judge` skill from the pmstack plugin with these arguments: $ARGUMENTS. If the skill is not available, tell the user to run /plugin install pmstack@pmstack.
Alternatives in Testing
- Loki Spec Status — Check whether the spec has drifted from its lock using Loki's living-spec drift detection, and summarize the r 1.1k ★
- Adversarial Spec — A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until a 514 ★
- Schematic — Claude Code / Codex skill: reverse engineer a product & technical spec from a git branch 150 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Desk Test
Backtest a structure with modelled premiums and judge honestly whether the result means anything
Claude Interaction Tests
Test whether your CLAUDE.md actually works. A Claude Code skill that restructures agent-facing docs (CLAUDE.md
Think Stress Test Decision
Pressure-test a consequential decision before committing, by comparing options, surfacing the conditions that
Speckit Constitution
Project kickoff for spec-driven work. Interview the user, then write the constitution (CLAUDE.md) and the prod
Test Login And Upload
End-to-end test of the CLI on the dev environment. Run each step sequentially — stop and report if any step fa
Test Upload Evals Skill
End-to-end test of publishing a skill with evals on the dev environment. Run each step sequentially — stop and
Related Agents
Test Archaeologist
Writes tests for legacy code by first excavating its actual behavior contract — what the code really does, not
TDD Test Writer
Writes failing tests for a ledger item. NEVER writes production code. Reads the spec, writes the test, runs it
Decide Well
Structured decision-making for high-stakes, often irreversible product choices. Use when facing a decision tha
