> ## Documentation Index
> Fetch the complete documentation index at: https://docs.contenthero.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Skills Overview

> Installable agent skills that encode the ContentHero workflows and our model-specific prompt craft.

The ContentHero skills are installable workflows for your agent. They turn ContentHero into a context-and-execution layer your agent already knows how to use well: how to ground a draft, how to chain a generation, how to prompt each model, how to assemble and publish a post. They run inside your agent (Claude Code, Cursor, Codex, and more) and drive the MCP, CLI, or raw API underneath.

The repo is [`contenthero-ai/skills`](https://github.com/contenthero-ai/skills).

## The three skills

<CardGroup cols={3}>
  <Card title="Generate" icon="wand-magic-sparkles" href="/skills/generate">
    Idea to media: image, video, audio, board, and lip-sync, with chaining and cost preflight.
  </Card>

  <Card title="Pipeline" icon="diagram-project" href="/skills/pipeline">
    The hero. Research to published: ground, draft in your voice, produce, assemble, schedule.
  </Card>

  <Card title="Brand" icon="book" href="/skills/brand">
    Your context. Read and grow your brand kit and knowledge base; feed the other two.
  </Card>
</CardGroup>

They cooperate: `contenthero-brand` resolves your brand context, `contenthero-pipeline` grounds and publishes, and `contenthero-generate` produces the media along the way. Most people install all three.

## What the skills add

ContentHero supplies the tools; the skills supply the craft of using them well:

* **Model prompt craft** specific to our roster, so a prompt is tuned to the model that will run it.
* **Grounded voice-synthesis**, the method for drafting copy that sounds like you (see [Voice synthesis](/skills/voice-synthesis)).
* **The workflow**, so the agent knows the order: ground, draft, approve, produce, assemble, publish.

When your agent drafts copy, the skills have it ground the draft in your real context (your outliers, brand voice, and past posts) and get your approval before publishing.

## The frozen-snapshot model

An installed skill is a git-clone snapshot. It does not auto-update; you update it with `git pull` or by re-running the installer. The skill is only instructions. The services it drives (the hosted MCP, the CLI to `/api/v1`, the model roster, pricing) are server-side and always live. So the skills bake durable guidance (craft and workflow) and defer volatile facts (model ids, prices) to live calls, which keeps a year-old install correct.

This is also why MCP ranks first in the [auth ladder](/for-ai-agents): it is the only self-updating transport.

## Install

<Card title="Install the skills" icon="download" href="/skills/install">
  Per-host install, the paste-prompt one-liner, and how to authenticate.
</Card>