# claude-repl

`station__claude-ai__claude-repl` · external (needs EXECUTION_BACKEND_URL configured) · domain `claude-ai` · pv-relevance `non-pv`

Bridge to Claude Code CLI.

> **Note:** This tool routes through an external execution backend. If `EXECUTION_BACKEND_URL` is unset on the server, calls return JSON-RPC error `-32603 "Tool execution backend not configured"`. Tools with `backend: native` execute in-process and are always callable.

## Agent metadata

- `idempotent`: unknown
- `read_only`: unknown
- `expected_latency_ms`: unknown (not yet contract-tested)
- `cost_tokens_estimate`: unknown

## Input schema

- `allowed_tools` *array* — Allowed tools list.
- `append_system_prompt` *string* — System prompt append.
- `max_output_bytes` *integer* — Maximum output bytes.
- `mcp_config_path` *string* — Optional MCP config path.
- `model` *string* — Optional model (e.g., "sonnet").
- `output_format` *string* — Output format.
- `permission_mode` *string* — Permission mode.
- `persist_session` *boolean* — Persist session boolean.
- `prompt` *string* (required) — The prompt to send.
- `session_id` *string* — Optional session ID.
- `settings_path` *string* — Optional settings path.
- `strict_mcp_config` *boolean* — Validate MCP config boolean.
- `system_prompt` *string* — System prompt override.
- `timeout_ms` *integer* — Timeout in ms.

## Example call

```json
POST /api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "station__claude-ai__claude-repl",
    "arguments": {
      "prompt": ""
    }
  }
}
```

## Related

- [/tools](/tools) — all 3062 tools
- [/tools/claude-ai__claude-repl](/tools/claude-ai__claude-repl) — HTML page
- [/tools/claude-ai__claude-repl/json](/tools/claude-ai__claude-repl/json) — JSON form (agent-friendly)
- [/api/mcp](/api/mcp) — endpoint
- [/AGENTS.md](/AGENTS.md) — agent guide