Eval Suite — Testing skill for Claude Code
Re-score tiny-spec against the SDD evaluation rubric, record it, and report the delta vs the last run.
How to install Eval Suite
Installs to ~/.claude/commands/grayma77er-tiny-spec-eval-suite.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/GrayMa77er/tiny-spec/HEAD/.claude/commands/eval-suite.md -o ~/.claude/commands/grayma77er-tiny-spec-eval-suite.md Restart Claude Code, or start a new session, for it to be picked up.
What Eval Suite does
description: Re-score tiny-spec against the SDD evaluation rubric, record it, and report the delta vs the last run. allowed-tools: Read, Grep, Glob, Bash, Edit, Write, Agent
/eval-suite — self-evaluate tiny-spec and track the score over time
You are scoring **this repo's tiny-spec suite** against the rubric in `docs/sdd-evaluation-rubric.md`, recording the result, and reflecting on how recent changes moved the score. This is a **maintainer dev tool** — it is *not* part of the install
Alternatives in Testing
- Marketing Skills Versions — Current versions of all skills 13.8k ★
- Cc Sdd — Spec-driven development (SDD) for your team's workflow 3.1k ★
- AI Contribution Skills (Engram) — branch-pr: clean branch and PR workflow 2.5k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Sdd Reconcile.Prompt
Fold already-shipped, undocumented code changes back into a baseline domain's spec as a retroactive, diff-wind
Agent Eval Design
Design an evaluation for an AI agent or LLM feature: what to test, how to grade it, and how to catch regressio
AI Evo
Self-hosted AI evaluation platform for LLMs & Agents — 9 eval types with AI judges, datasets, pre-eval scripts
Agentic QA Copilot
Automated call center QA analytics agent — scores every call against a weighted rubric (with emphasis on ident
Gov Eval
Run GovEval — the governance validation test suite.
Superpowers Evals
Behavioral eval lab (Quorum) for the superpowers project that drives real coding-agent CLIs (Claude, Codex, Ge
Related Agents
ADR Critic
Use when a drafted Architecture Decision Record must be checked against a written standard before it is accept
Sdd Reconcile.Agent
Fold already-shipped, undocumented code changes back into a baseline domain spec as a diff-window-scoped retro
Bench Reviewer
Reviews eval suite quality — fixture/rubric consistency, scoring accuracy, false positive traps, difficulty ca