W&B Skills for Claude Code, Codex, and other coding agents logo

SKILLS

W&B Skills for Claude Code, Codex, and other coding agents

by Weights & Biases

Weights & BiasesW&Bagent skills

Overview

Official W&B agent skills for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and other coding agents.

Details

W&B Skills are official reusable instruction sets from Weights & Biases for coding agents. The W&B documentation describes them as skills for agents including Claude Code, Codex, Cursor, GitHub Copilot, and Gemini CLI, installable with commands that use wandb/skills. Source descriptions indicate they help agents monitor and compare W&B runs, spot training issues, and review ML experiment performance using the Weights & Biases AI developer platform, including Models and Weave workflows.

When to Use

Use when a coding agent needs project-specific guidance for working with Weights & Biases experiment tracking model runs or Weave-related workflows. Use to help Claude Code Codex Cursor GitHub Copilot Gemini CLI or similar coding-agent CLIs inspect W&B runs compare experiment performance and identify training issues.

Getting Started

  1. Open the official W&B Skills documentation at https://docs.wandb.ai/platform/wb-skills.
  2. Review the official GitHub repository at https://github.com/wandb/skills for available skill files and installation instructions.
  3. Install the relevant skills using the documented commands that reference wandb/skills
  4. then test them in a supported coding agent such as Claude Code
  5. Codex
  6. Cursor
  7. GitHub Copilot
  8. or Gemini CLI.

Key Features

  • Reusable instruction sets for coding agents working with Weights & Biases.
  • Official wandb/skills GitHub repository for W&B Models and Weave agent skills.
  • Documented support for multiple coding agents
  • including Claude Code
  • Codex
  • Cursor
  • GitHub Copilot
  • and Gemini CLI.
  • Agent workflows for monitoring W&B runs
  • comparing runs
  • spotting training issues
  • and reviewing ML experiment performance.

Capabilities

  • coding-agent-skill
  • ml-experiment-analysis
  • wandb-run-monitoring
  • training-issue-review

Last updated Jul 29, 2026