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

# Claude Code

> Run agents on Claude Code: models, capabilities, tools, and what paddock configures for you.

Claude Code is the strongest harness on Anthropic models, with skills, hooks, plugins, subagents, workflows, and MCP. paddock runs the real Claude Code CLI, unmodified, at the version it pins (currently **2.1.300**), in each agent's sandbox.

```yaml theme={null}
spec:
  harness: claude-code
  model: claude-opus
  small_model: claude-haiku
```

## Models

Claude Code runs Anthropic models over the Anthropic Messages API:

* Platform models: `claude-opus`, `claude-sonnet`, and `claude-haiku`.
* Project models of an `anthropic` or `anthropic-compatible` provider, named `provider/model`. See [Models](https://docs.paddock.build/guides/models.md).

Claude Code has two model settings, chosen independently from either source:

| Setting | Used for | Changes reach |
| - | - | - |
| `model` | The conversation | Running sessions, between turns |
| `small_model` | Background requests, and subagents that ask for a smaller model; default `claude-haiku` | Sessions created afterwards |

## Capabilities

| Capability | Supported |
| - | - |
| MCP servers, remote and local | Yes |
| Marketplaces and plugins, with their skills, subagents, hooks, and MCP servers | Yes |
| Workflows | Yes: listed read-only by `agent.workflows`, and shown with their runs in the harness status |
| Memory | Yes |

A plugin is installed with Claude Code's own `claude plugin` commands and taken up by the running process without a restart; an MCP server with `claude mcp`. Changes apply between turns; see [Capabilities](https://docs.paddock.build/guides/capabilities.md).

## Configuration

| Setting | On Claude Code |
| - | - |
| System prompt | Appended to Claude Code's own system prompt (`--append-system-prompt`). Claude Code records a conversation's system prompt on its first request and reuses it on every resume, which is why a session's system prompt is fixed at its creation |
| Permissions | The `allow` and `deny` lists of Claude Code's settings, as given, in Claude Code's permission rule syntax, such as `Bash(npm run test:*)` |
| Environment | The `env` of Claude Code's settings |
| Memory | Rendered into the session's `CLAUDE.md`, read when the harness process starts |

The harness runs non-interactively: there is no one to answer an approval prompt.

## Tools

Claude Code's tools map to paddock's typed tools:

| paddock tool | Claude Code tool |
| - | - |
| `shell` | `Bash` |
| `file_read` | `Read` |
| `file_write` | `Write` |
| `file_edit` | `Edit` |
| `glob`, `grep` | `Glob`, `Grep`, when Claude Code's settings enable them; by default it searches with `Bash` |
| `web_fetch`, `web_search` | `WebFetch`, `WebSearch` |
| `mcp` | `mcp__<server>__<tool>` |
| `subagent` | `Agent` (or `Task`) |
| `skill` | `Skill`, with the skill's name, such as `frontend-design:frontend-design` for a plugin's skill |
| `unknown` | Any other tool, such as its task tools, with the raw input |

Reasoning arrives as `reasoning` messages, and items paddock does not map, such as a context compaction boundary, as `unknown` messages with their raw JSON.

## Harness status

Claude Code reports every section of the [harness status](https://docs.paddock.build/guides/harness-status.md). Its Context section is what `/context` shows: usage by category, counted with the model endpoint's token counting, and the MCP tools, custom agents, memory files, and skills that take space.

## Background work

Claude Code keeps background shells, subagents, and workflows running after a turn ends. When one finishes, Claude Code may start a turn on its own: paddock records it as a **harness turn**, which consumes no message. Stopping a session ends the current turn but spares background work. An agent does not sleep while a background agent or workflow runs.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.