Agent skills for integrating ZeroEval into AI applications. Works with Cursor, Claude Code, Codex, and 30+ other coding agents.
zeroeval-install -- Get from zero to production-ready. Walks through SDK install (Python or TypeScript), first trace, ze.prompt migration, and starter judge recommendations. Routes to custom-tracing for non-SDK languages.
custom-tracing -- Send traces to ZeroEval without the SDK. Covers direct REST API ingestion (POST /spans) and OpenTelemetry (OTLP) export for any language.
create-judge -- Design and create automated judges. Covers binary vs scored evaluation, template writing, criteria design, and the full creation API.
prompt-migration -- Migrate hardcoded prompts to ze.prompt. Covers the full migration workflow, feedback wiring, judge linkage, staged rollout, and prompt optimization for both Python and TypeScript.
| I want to... | Use |
|---|---|
| Add ZeroEval to my project for the first time | zeroeval-install |
| Set up tracing and observability (Python or TypeScript) | zeroeval-install |
| Send traces without the SDK (Go, Ruby, Java, etc.) | custom-tracing |
| Send traces via REST API or OpenTelemetry (OTLP) | custom-tracing |
| Migrate hardcoded prompts to ze.prompt | prompt-migration |
| Wire feedback collection for prompt optimization | prompt-migration |
| Connect judges to a prompt for automated evaluation | prompt-migration |
| Understand the staged rollout (explicit / auto / latest) | prompt-migration |
| Get recommendations for which judges to create | zeroeval-install |
| Create a new judge with a custom template | create-judge |
| Design scoring criteria for multi-dimensional evaluation | create-judge |
| Understand the judge creation API | create-judge |
# Install all skills
npx skills add zeroeval/zeroeval-skills
# Install a specific skill
npx skills add zeroeval/zeroeval-skills --skill zeroeval-install
# List available skills
npx skills add zeroeval/zeroeval-skills --list# Add the marketplace
/plugin marketplace add zeroeval/zeroeval-skills
# Install a specific plugin
/plugin install zeroeval-install@zeroeval-skills
/plugin install custom-tracing@zeroeval-skills
/plugin install create-judge@zeroeval-skills
/plugin install prompt-migration@zeroeval-skills
# Reload plugins if the new commands do not appear immediately
/reload-pluginsClaude Code plugin skills are namespaced by plugin name. After installing from the marketplace, invoke them as:
/zeroeval-install:zeroeval-install
/custom-tracing:custom-tracing
/create-judge:create-judge
/prompt-migration:prompt-migrationgit clone https://github.com/zeroeval/zeroeval-skills.git
mkdir -p .agents/skills
cp -r zeroeval-skills/skills/* .agents/skills/Or copy individual skills:
# Cursor
mkdir -p .cursor/skills
cp -r zeroeval-skills/skills/zeroeval-install .cursor/skills/zeroeval-install
cp -r zeroeval-skills/skills/custom-tracing .cursor/skills/custom-tracing
cp -r zeroeval-skills/skills/create-judge .cursor/skills/create-judge
cp -r zeroeval-skills/skills/prompt-migration .cursor/skills/prompt-migration
# Claude Code
mkdir -p .claude/skills
cp -r zeroeval-skills/skills/zeroeval-install .claude/skills/zeroeval-install
cp -r zeroeval-skills/skills/custom-tracing .claude/skills/custom-tracing
cp -r zeroeval-skills/skills/create-judge .claude/skills/create-judge
cp -r zeroeval-skills/skills/prompt-migration .claude/skills/prompt-migrationNote: on Windows without symlink support, use npx skills or the manual copy method. The plugins/ directory contains symlinks that may not resolve on Windows.
zeroeval-skills/
├── .claude-plugin/
│ └── marketplace.json # Claude Code plugin marketplace
├── README.md
├── skills/ # Canonical skill content (npx skills + manual copy)
│ ├── zeroeval-install/
│ │ ├── SKILL.md
│ │ └── references/
│ │ ├── python-integration-playbook.md
│ │ ├── typescript-integration-playbook.md
│ │ ├── judges-playbook.md
│ │ └── troubleshooting.md
│ ├── custom-tracing/
│ │ ├── SKILL.md
│ │ └── references/
│ │ └── api-integration-playbook.md
│ ├── create-judge/
│ │ ├── SKILL.md
│ │ └── references/
│ │ ├── judge-creation-guide.md
│ │ ├── evaluation-patterns.md
│ │ └── defaults-and-api-reference.md
│ └── prompt-migration/
│ ├── SKILL.md
│ └── references/
│ ├── python-prompt-migration-playbook.md
│ └── typescript-prompt-migration-playbook.md
└── plugins/ # Claude Code plugin wrappers (symlinked)
├── zeroeval-install/
├── custom-tracing/
├── create-judge/
└── prompt-migration/
- A ZeroEval account and API key -- zeroeval.com
- SDK path: Python 3.8+ or Node 18+, plus an LLM provider SDK (OpenAI, Vercel AI, LangChain, etc.)
- Direct API / OTLP path: any language with an HTTP client or OpenTelemetry exporter