Anthropic

Use Claude Sonnet 4, Claude Sonnet 4.5, and other Anthropic models with docker-agent.

Setup

# Set your API key
export ANTHROPIC_API_KEY="sk-ant-..."

Workload Identity Federation (no API key)

Authenticate with short-lived tokens minted from your own OIDC identity provider instead of a long-lived API key. See Anthropic’s Workload Identity Federation guide to provision a Federation Rule, then configure docker-agent with a typed auth: block:

providers:
  anthropic-wif:
    provider: anthropic
    auth:
      type: workload_identity_federation
      workload_identity_federation:
        federation_rule_id: fdrl_REPLACE_ME
        organization_id: 00000000-0000-0000-0000-000000000000
        # Optional: only required for target_type=SERVICE_ACCOUNT rules.
        service_account_id: svac_REPLACE_ME
        identity_token:
          # Pick exactly one of: file, env, command, url
          file: /var/run/secrets/anthropic.com/token

models:
  claude:
    provider: anthropic-wif
    model: claude-sonnet-4-5

identity_token accepts four mutually exclusive sources:

For url, both the URL and any header values support ${VAR} expansion against the runtime environment, which lets you wire the GitHub Actions OIDC token endpoint without putting secrets in YAML:

identity_token:
  url: ${ACTIONS_ID_TOKEN_REQUEST_URL}&audience=https://api.anthropic.com
  headers:
    Authorization: bearer ${ACTIONS_ID_TOKEN_REQUEST_TOKEN}
  response_field: value

auth: is mutually exclusive with --gateway. Token-refresh failures are surfaced through the normal error path with a clear anthropic workload identity federation: failed to refresh identity token from <kind> source (federation_rule=fdrl_…): ... message in the TUI.

A complete walkthrough of all four sources lives in examples/anthropic_wif.yaml.

Configuration

Inline

agents:
  root:
    model: anthropic/claude-sonnet-4-5

Named Model

models:
  claude:
    provider: anthropic
    model: claude-sonnet-4-5
    max_tokens: 64000

Available Models

Thinking Budget

Anthropic uses integer token budgets (1024–32768). Defaults to 8192 with interleaved thinking enabled:

models:
  claude-deep:
    provider: anthropic
    model: claude-sonnet-4-5
    thinking_budget: 16384 # must be < max_tokens

Interleaved Thinking

Enabled by default. Allows tool calls during model reasoning for more integrated problem-solving:

models:
  claude:
    provider: anthropic
    model: claude-sonnet-4-5
    provider_opts:
      interleaved_thinking: false # disable if needed

Task Budget

task_budget caps the total number of tokens the model may spend across a multi-step agentic task — combined thinking, tool calls, and final output. It is forwarded as output_config.task_budget and is ideal for letting long-running agents self-regulate effort without tightening max_tokens on every call.

docker-agent automatically attaches the required task-budgets-2026-03-13 beta header whenever this field is set. You can configure task_budget on any Claude model — docker-agent never gates it by model name. At the time of writing, only Claude Opus 4.7 actually honors the field; other Claude models (Sonnet 4.5, Opus 4.5 / 4.6, etc.) are expected to reject requests that include it. Check the Anthropic release notes linked above for the current list of supported models.

models:
  opus:
    provider: anthropic
    model: claude-opus-4-7
    task_budget: 128000 # integer shorthand → { type: tokens, total: 128000 }
    thinking_budget: adaptive

Object form (forward-compatible with future budget types):

  opus:
    provider: anthropic
    model: claude-opus-4-7
    task_budget:
      type: tokens
      total: 128000

See the full schema on the Model Configuration page.

Thinking Display

Controls whether thinking blocks are returned in responses when thinking is enabled. Claude Opus 4.7 hides thinking content by default (omitted); earlier Claude 4 models default to summarized. Set thinking_display in provider_opts to override:

models:
  claude-opus-4-7:
    provider: anthropic
    model: claude-opus-4-7
    thinking_budget: adaptive
    provider_opts:
      thinking_display: summarized # "summarized", "display", or "omitted"

Valid values:

• summarized: thinking blocks are returned with summarized thinking text (default for Claude 4 models prior to Opus 4.7).
• display: thinking blocks are returned for display (use this to re-enable thinking output on Opus 4.7).
• omitted: thinking blocks are returned with an empty thinking field; the signature is still returned for multi-turn continuity (default for Opus 4.7). Useful to reduce time-to-first-text-token when streaming.

Note: thinking_display applies to both thinking_budget with token counts and adaptive/effort-based budgets. Full thinking tokens are billed regardless of the thinking_display value.