Adaline Skills

Skills that guide AI coding agents to integrate with the Adaline platform — send traces, manage prompts, run evaluations, fetch deployments, and more. Compatible with Cursor, Claude Code, Codex, Windsurf, and 40+ other agents.

Install

npx (recommended)

npx skills add adaline/skills

Install a single skill:

npx skills add adaline/skills --skill adaline-logs

Preview available skills:

npx skills add adaline/skills --list

Install for specific agents:

npx skills add adaline/skills -a claude-code -a cursor -y

The CLI auto-detects installed agents and symlinks skills into the correct directory (.claude/skills/, .cursor/skills/, etc.).

Claude Code plugin

/plugin marketplace add adaline/skills
/plugin install skills

Manual

Clone the repo and copy skill folders into your agent's skills directory:

git clone https://github.com/adaline/skills.git

Then copy the skills you need:

  • Claude Code: .claude/skills/
  • Cursor: .cursor/skills/
  • Codex: .codex/skills/

Update

npx skills check     # check for updates
npx skills update    # update all installed skills

Configuration

When you have your Adaline credentials, set these environment variables:

export ADALINE_API_KEY=your-api-key           # from Settings > API Keys in the dashboard
export ADALINE_PROJECT_ID=your-project-id     # from the project sidebar in the dashboard

No credentials yet? Skills use placeholder values by default — you can start integrating now and replace them with real values once your account is ready. Sign up at https://app.adaline.ai/sign-up

Available Skills

Skill What it does
adaline-integration High-level integration guide. Start here — helps you choose the right skills and approach (API, TypeScript SDK, Python SDK) for your setup.
adaline-logs Send traces and spans to Adaline for AI agent observability. Covers REST API, TypeScript SDK, and Python SDK instrumentation.
adaline-deployments Fetch deployed prompt snapshots at runtime. Use environment-based latest deployment lookup or pin a concrete deployment ID.
adaline-prompts Create and manage prompts programmatically via the REST API.
adaline-datasets Build and manage evaluation datasets — columns, rows, CSV import, dynamic columns.
adaline-evaluators Create evaluators: LLM-as-a-judge, JavaScript, text-matcher, cost, latency, response-length.
adaline-evaluations Run evaluations at scale, check status, analyze per-row results, cancel runs.
adaline-providers List configured AI providers (OpenAI, Anthropic, Google, etc.) and available models.

Start Here

Use adaline-integration first — it helps you choose which skills to adopt based on your runtime, codebase, and goals.

Recommended order:

  1. adaline-logs — get observability first
  2. adaline-deployments — fetch the active prompt snapshot in your app
  3. adaline-datasets + adaline-evaluators + adaline-evaluations — quality gates
  4. adaline-providers — discover available models

Integration Approaches

Approach Best for Skills that support it
TypeScript SDK (@adaline/client) Node.js/TypeScript apps adaline-logs, adaline-deployments
Python SDK (adaline-client) Python apps adaline-logs, adaline-deployments
REST API Any language, serverless All skills
Adaline Proxy (zero-code) Quick start, simple apps adaline-logs (see SKILL.md)

Repository structure

skills/
  adaline-integration/        # Master integration guide
  adaline-logs/               # Traces and spans
  adaline-deployments/        # Deployment fetching
  adaline-prompts/            # Prompt management
  adaline-datasets/           # Dataset management
  adaline-evaluators/         # Evaluator creation
  adaline-evaluations/        # Running evaluations
  adaline-providers/          # Provider/model discovery

scripts/                      # CI validation
.claude-plugin/               # Plugin config for distribution

Links