Skip to main content
Global flags (available on all commands):
  • -t, --token <TOKEN>: require/use bearer auth
  • -n, --no-token: disable auth

​
server

Run the HTTP server. 
sandbox-agent server [OPTIONS]
OptionDefaultDescription
-H, --host <HOST>127.0.0.1Host to bind
-p, --port <PORT>2468Port to bind
-O, --cors-allow-origin <ORIGIN>-Allowed CORS origin (repeatable)
-M, --cors-allow-method <METHOD>allAllowed CORS method (repeatable)
-A, --cors-allow-header <HEADER>allAllowed CORS header (repeatable)
-C, --cors-allow-credentialsfalseEnable CORS credentials
--no-telemetryfalseDisable anonymous telemetry
sandbox-agent server --port 3000
Notes:
  • Server logs are redirected to files by default.
  • Set SANDBOX_AGENT_LOG_STDOUT=1 to force stdout/stderr logging.
  • Use SANDBOX_AGENT_LOG_DIR to override log directory.

​
install-agent

Install or reinstall a single agent. 
sandbox-agent install-agent <AGENT> [OPTIONS]
OptionDescription
-r, --reinstallForce reinstall
--agent-version <VERSION>Override agent package version
--agent-process-version <VERSION>Override agent process version
sandbox-agent install-agent claude --reinstall

​
opencode (experimental)

Start/reuse daemon and run opencode attach against /opencode. 
sandbox-agent opencode [OPTIONS]
OptionDefaultDescription
-H, --host <HOST>127.0.0.1Daemon host
-p, --port <PORT>2468Daemon port
--session-title <TITLE>-Reserved option (currently no-op)
--yolofalseOpenCode attach mode flag
sandbox-agent opencode

​
daemon

Manage the background daemon.

​
daemon start

sandbox-agent daemon start [OPTIONS]
OptionDefaultDescription
-H, --host <HOST>127.0.0.1Host
-p, --port <PORT>2468Port
--upgradefalseUse ensure-running + upgrade behavior
sandbox-agent daemon start
sandbox-agent daemon start --upgrade

​
daemon stop

sandbox-agent daemon stop [OPTIONS]
OptionDefaultDescription
-H, --host <HOST>127.0.0.1Host
-p, --port <PORT>2468Port

​
daemon status

sandbox-agent daemon status [OPTIONS]
OptionDefaultDescription
-H, --host <HOST>127.0.0.1Host
-p, --port <PORT>2468Port

​
credentials

​
credentials extract

sandbox-agent credentials extract [OPTIONS]
OptionDescription
-a, --agent <AGENT>Filter by claude, codex, opencode, or amp
-p, --provider <PROVIDER>Filter by provider
-d, --home-dir <DIR>Override home dir
--no-oauthSkip OAuth sources
-r, --revealShow full credential values
sandbox-agent credentials extract --agent claude --reveal

​
credentials extract-env

sandbox-agent credentials extract-env [OPTIONS]
OptionDescription
-e, --exportPrefix output with export
-d, --home-dir <DIR>Override home dir
--no-oauthSkip OAuth sources
eval "$(sandbox-agent credentials extract-env --export)"

​
api

API subcommands for scripting. Shared option:
OptionDefaultDescription
-e, --endpoint <URL>http://127.0.0.1:2468Target server

​
api agents

sandbox-agent api agents list [--endpoint <URL>]
sandbox-agent api agents report [--endpoint <URL>]
sandbox-agent api agents install <AGENT> [--reinstall] [--endpoint <URL>]

​
api agents list

List all agents and their install status. 
sandbox-agent api agents list

​
api agents report

Emit a JSON report of available models, modes, and thought levels for every agent. Calls GET /v1/agents?config=true and groups each agent’s config options by category. 
sandbox-agent api agents report --endpoint http://127.0.0.1:2468 | jq .
Example output: 
{
  "generatedAtMs": 1740000000000,
  "endpoint": "http://127.0.0.1:2468",
  "agents": [
    {
      "id": "claude",
      "installed": true,
      "models": {
        "currentValue": "default",
        "values": [
          { "value": "default", "name": "Default" },
          { "value": "sonnet", "name": "Sonnet" },
          { "value": "opus", "name": "Opus" },
          { "value": "haiku", "name": "Haiku" }
        ]
      },
      "modes": {
        "currentValue": "default",
        "values": [
          { "value": "default", "name": "Default" },
          { "value": "acceptEdits", "name": "Accept Edits" },
          { "value": "plan", "name": "Plan" },
          { "value": "dontAsk", "name": "Don't Ask" },
          { "value": "bypassPermissions", "name": "Bypass Permissions" }
        ]
      },
      "thoughtLevels": { "values": [] }
    }
  ]
}
See Agent Capabilities for a full reference of supported models, modes, and thought levels per agent.

​
api agents install

sandbox-agent api agents install codex --reinstall