Agent Skills

Updated Oct 28, 2026

Fairspec ships Agent Skills that tell coding agents such as Claude Code, Cursor and Codex how to work with Fairspec data. Once installed, your agent knows which command, MCP tool or Python call fits a task, and how to read the report it gets back. You don't have to explain any of it in the prompt.

Skill	Your agent uses it when
`fairspec-exploration`	previewing, describing or querying data
`fairspec-inference`	describing files as dialects, schemas, datasets
`fairspec-validation`	checking data or schemas and fixing the errors

Install from the website

fairspec.org publishes its skills under /.well-known/agent-skills/, so the skills CLI can install them from the site URL. Run this at your project root:

npx skills add https://fairspec.org

It asks which skills and agents to install for. To skip the prompts, pass them as flags:

npx skills add https://fairspec.org --skill fairspec-validation --agent claude-code

Add --global to install for your user instead of the project, and --list to see what's available without installing anything. Run npx skills update later to pull newer versions.

Use them

You don't need to invoke a skill yourself. Agents read each skill's description and load it when a task matches: asking to check a CSV against its schema loads fairspec-validation, and asking to describe a folder of files loads fairspec-inference. If your agent supports invoking skills by name, you can also call one directly, as with /fairspec-validation in Claude Code.

Each skill covers every way to run a task: the terminal, the MCP server and the Python library. For example, validating a table with the library:

from fairspec import Resource, validate_table

resource = Resource(data="broken.csv", tableSchema="schema.json")
report = validate_table(resource)

print(report.valid)
# False