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

# Skills

> List, install, and manage playbooks that teach an AI agent which tool to use, bundled with the Zenrows CLI.

A skill is a short, agent-readable playbook. Install one and an AI agent knows when and how to use a Zenrows tool, no extra prompting needed.

```bash theme={null}
zenrows skill list
zenrows skill install --all
```

| Command                                      | Does                                    |
| -------------------------------------------- | --------------------------------------- |
| `zenrows skill list`                         | List all skills, installed or not       |
| `zenrows skill install <name>`               | Install a skill                         |
| `zenrows skill install --all`                | Install every skill                     |
| `zenrows skill explain <name>`               | Show a skill's full docs                |
| `zenrows skill validate <name>`              | Check an installed skill is well-formed |
| `zenrows skill generate --from-run <run-id>` | Scaffold a new skill from a past run    |
| `zenrows skill update [name]`                | Refresh installed skill(s)              |
| `zenrows skill remove <name>`                | Remove an installed skill               |

## Catalog

| Skill               | Description                                                        |
| ------------------- | ------------------------------------------------------------------ |
| `zenrows`           | Master playbook: the decision tree, credit costs, escalation rules |
| `protected-fetch`   | Use Protected Fetch for anti-bot-protected page retrieval          |
| `extract`           | Turn protected pages into structured data                          |
| `batch-jobs`        | Scale fetch/extract over many URLs via Batch                       |
| `interact-browser`  | Escalate to a browser only when fetch/extract can't do the job     |
| `cost-control`      | Pick the cheapest reliable option; understand the credit costs     |
| `trace-debug`       | Read run traces and choose the next action after a failure         |
| `compliance-policy` | Domain allow/deny rules, credit limits, secret handling            |