Claude Agent SDK for Elixir

An Elixir SDK aiming for feature-complete parity with the official claude-agent-sdk-python. Build AI-powered applications with Claude using a production-ready interface for the Claude Code CLI, featuring streaming responses, lifecycle hooks, permission controls, and in-process tool execution via MCP.

Note: This SDK does not bundle the Claude Code CLI. You must install it separately (see Prerequisites).

What You Can Build

• AI coding assistants with real-time streaming output
• Automated code review pipelines with custom permission policies
• Multi-agent workflows with specialized personas
• Tool-augmented applications using the Model Context Protocol (MCP)
• Interactive chat interfaces with typewriter-style output

Installation

Add to your mix.exs:

def deps do
  [
    {:claude_agent_sdk, "~> 0.11.0"}
  ]
end

Then fetch dependencies:

mix deps.get

Prerequisites

Install the Claude Code CLI (requires Node.js):

npm install -g @anthropic-ai/claude-code

Verify installation:

claude --version

Quick Start

1. Authenticate

Choose one method:

# Option A: Environment variable (recommended for CI/CD)
export ANTHROPIC_API_KEY="sk-ant-api03-..."

# Option B: OAuth token
export CLAUDE_AGENT_OAUTH_TOKEN="sk-ant-oat01-..."

# Option C: Interactive login
claude login

2. Run Your First Query

alias ClaudeAgentSDK.{ContentExtractor, Options}

# Simple query with streaming collection
ClaudeAgentSDK.query("Write a function that calculates factorial in Elixir")
|> Enum.each(fn msg ->
  case msg.type do
    :assistant -> IO.puts(ContentExtractor.extract_text(msg) || "")
    :result -> IO.puts("Done! Cost: $#{msg.data.total_cost_usd}")
    _ -> :ok
  end
end)

3. Real-Time Streaming

alias ClaudeAgentSDK.Streaming

{:ok, session} = Streaming.start_session()

Streaming.send_message(session, "Explain GenServers in one paragraph")
|> Stream.each(fn
  %{type: :text_delta, text: chunk} -> IO.write(chunk)
  %{type: :message_stop} -> IO.puts("")
  _ -> :ok
end)
|> Stream.run()

Streaming.close_session(session)

If session initialization or message send fails, the stream now emits an immediate %{type: :error, error: reason} event instead of waiting for the 5-minute stream timeout.

Authentication

The SDK supports three authentication methods, checked in this order:

Cloud Providers

AWS Bedrock:

export CLAUDE_AGENT_USE_BEDROCK=1
export AWS_ACCESS_KEY_ID=...
export AWS_SECRET_ACCESS_KEY=...
export AWS_REGION=us-west-2

Google Vertex AI:

export CLAUDE_AGENT_USE_VERTEX=1
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/key.json
export GOOGLE_CLOUD_PROJECT=your-project-id

Token Setup (Local Development)

For persistent authentication without re-login:

mix claude.setup_token

AuthManager keeps running if token storage save/clear fails and returns {:error, reason}. Handle clear_auth/0 accordingly in your app code:

case ClaudeAgentSDK.AuthManager.clear_auth() do
  :ok -> :ok
  {:error, reason} -> IO.puts("Failed to clear auth: #{inspect(reason)}")
end

Check authentication status:

alias ClaudeAgentSDK.AuthChecker
diagnosis = AuthChecker.diagnose()
# => %{authenticated: true, auth_method: "Anthropic API", ...}

Core Concepts

Choosing the Right API

Query API

The simplest way to interact with Claude:

# Basic query
messages = ClaudeAgentSDK.query("What is recursion?") |> Enum.to_list()

# With options
opts = %ClaudeAgentSDK.Options{
  model: "sonnet",
  max_turns: 5,
  output_format: :stream_json
}
messages = ClaudeAgentSDK.query("Explain OTP", opts) |> Enum.to_list()

# Streamed input prompts (unidirectional)
prompts = [
  %{"type" => "user", "message" => %{"role" => "user", "content" => "Hello"}},
  %{"type" => "user", "message" => %{"role" => "user", "content" => "How are you?"}}
]

ClaudeAgentSDK.query(prompts, opts) |> Enum.to_list()

# Custom transport injection
ClaudeAgentSDK.query("Hello", opts, {ClaudeAgentSDK.Transport.Port, []})
|> Enum.to_list()

# Lazy transport startup (defer subprocess spawn to handle_continue)
ClaudeAgentSDK.query(
  "Hello",
  opts,
  {ClaudeAgentSDK.Transport.Port, [startup_mode: :lazy]}
)
|> Enum.to_list()

# Continue a conversation
ClaudeAgentSDK.continue("Can you give an example?") |> Enum.to_list()

# Resume a specific session
ClaudeAgentSDK.resume("session-id", "What about supervision trees?") |> Enum.to_list()

Streaming API

For real-time, character-by-character output:

alias ClaudeAgentSDK.{Options, Streaming}

{:ok, session} = Streaming.start_session(%Options{model: "haiku"})

# Send messages and stream responses
Streaming.send_message(session, "Write a haiku about Elixir")
|> Enum.each(fn
  %{type: :text_delta, text: t} -> IO.write(t)
  %{type: :tool_use_start, name: n} -> IO.puts("\nUsing tool: #{n}")
  %{type: :message_stop} -> IO.puts("\n---")
  _ -> :ok
end)

# Multi-turn conversation (context preserved)
Streaming.send_message(session, "Now write one about Phoenix")
|> Enum.to_list()

Streaming.close_session(session)

Subagent Streaming: When Claude spawns subagents via the Task tool, events include a parent_tool_use_id field to identify the source. Main agent events have nil, subagent events have the Task tool call ID. Streaming events also include uuid, session_id, and raw_event metadata for parity with the Python SDK. Stream event wrappers require uuid and session_id (missing keys raise). See the Streaming Guide for details.

Hooks System

Intercept and control agent behavior at key lifecycle points:

alias ClaudeAgentSDK.{Client, Options}
alias ClaudeAgentSDK.Hooks.{Matcher, Output}

# Block dangerous commands
check_bash = fn input, _id, _ctx ->
  case input do
    %{"tool_name" => "Bash", "tool_input" => %{"command" => cmd}} ->
      if String.contains?(cmd, "rm -rf") do
        Output.deny("Dangerous command blocked")
      else
        Output.allow()
      end
    _ -> %{}
  end
end

opts = %Options{
  hooks: %{
    pre_tool_use: [Matcher.new("Bash", [check_bash])]
  }
}

{:ok, client} = Client.start_link(opts)

Available Hook Events (all 12 Python SDK events supported):

• pre_tool_use / post_tool_use / post_tool_use_failure - Tool execution lifecycle
• user_prompt_submit - Before sending user messages
• stop / subagent_stop / subagent_start - Agent lifecycle
• notification - CLI notifications
• permission_request - Permission dialog interception
• session_start / session_end - Session lifecycle
• pre_compact - Before context compaction

See the Hooks Guide for comprehensive documentation.

Supervision

Hook and permission callbacks run in async tasks. For production, add the SDK task supervisor so callback processes are supervised:

children = [
  ClaudeAgentSDK.TaskSupervisor,
  {ClaudeAgentSDK.Client, options}
]

If you use a custom supervisor name, configure the SDK to match:

children = [
  {ClaudeAgentSDK.TaskSupervisor, name: MyApp.ClaudeTaskSupervisor}
]

config :claude_agent_sdk, task_supervisor: MyApp.ClaudeTaskSupervisor

If an explicitly configured supervisor is missing at runtime, the SDK logs a warning and falls back to Task.start/1. With default settings, missing ClaudeAgentSDK.TaskSupervisor falls back silently for backward compatibility. For stricter behavior in dev/test:

config :claude_agent_sdk, task_supervisor_strict: true

In strict mode, ClaudeAgentSDK.TaskSupervisor.start_child/2 returns {:error, {:task_supervisor_unavailable, supervisor}} instead of spawning an unsupervised fallback task.

Permission System

Fine-grained control over tool execution:

alias ClaudeAgentSDK.{Options, Permission.Result}

permission_callback = fn ctx ->
  case ctx.tool_name do
    "Write" ->
      # Redirect system file writes to safe location
      if String.starts_with?(ctx.tool_input["file_path"], "/etc/") do
        safe_path = "/tmp/sandbox/" <> Path.basename(ctx.tool_input["file_path"])
        Result.allow(updated_input: %{ctx.tool_input | "file_path" => safe_path})
      else
        Result.allow()
      end
    _ ->
      Result.allow()
  end
end

opts = %Options{
  can_use_tool: permission_callback,
  permission_mode: :default  # :default | :accept_edits | :plan | :bypass_permissions | :delegate | :dont_ask
}

Note: can_use_tool is mutually exclusive with permission_prompt_tool. The SDK routes can_use_tool through the control client (including string prompts), auto-enables include_partial_messages, and sets permission_prompt_tool to \"stdio\" internally so the CLI can emit permission callbacks. Use :default or :plan for built-in tool permissions; :delegate is intended for external tool execution. Hook-based fallback only applies in non-:delegate modes and ignores updated_permissions. If you do not see callbacks, your CLI build may not emit control callbacks (see examples/advanced_features/permissions_live.exs).

Stream a single client response until the final result:

Client.receive_response_stream(client)
|> Enum.to_list()

MCP Tools (In-Process)

Define custom tools that Claude can call directly in your application:

defmodule MyTools do
  use ClaudeAgentSDK.Tool

  deftool :calculate, "Perform a calculation", %{
    type: "object",
    properties: %{
      expression: %{type: "string", description: "Math expression to evaluate"}
    },
    required: ["expression"]
  } do
    def execute(%{"expression" => expr}) do
      # Your logic here
      result = eval_expression(expr)
      {:ok, %{"content" => [%{"type" => "text", "text" => "Result: #{result}"}]}}
    end
  end
end

# Create an MCP server with your tools
server = ClaudeAgentSDK.create_sdk_mcp_server(
  name: "calculator",
  version: "1.0.0",
  tools: [MyTools.Calculate]
)

# Optional: start tool registry under your DynamicSupervisor
{:ok, sup} = DynamicSupervisor.start_link(strategy: :one_for_one)

server = ClaudeAgentSDK.create_sdk_mcp_server(
  name: "calculator",
  version: "1.0.0",
  tools: [MyTools.Calculate],
  supervisor: sup
)

opts = %ClaudeAgentSDK.Options{
  mcp_servers: %{"calc" => server},
  allowed_tools: ["mcp__calc__calculate"]
}

Note: MCP server routing only supports initialize, tools/list, tools/call, and notifications/initialized. Calls to resources/list or prompts/list return JSON-RPC method-not-found errors to match the Python SDK. If version is omitted, it defaults to "1.0.0".

Configuration Options

Key options for ClaudeAgentSDK.Options:

CLI path override: set path_to_claude_code_executable or executable in Options (Python cli_path equivalent).

Runtime Application Config

config :claude_agent_sdk,
  # Timeout for in-process tool execution tasks in Tool.Registry
  tool_execution_timeout_ms: 30_000,
  # Query CLI stream backend module
  cli_stream_module: ClaudeAgentSDK.Query.CLIStream,
  # Fail fast when configured task supervisor is unavailable
  task_supervisor_strict: false

config :claude_agent_sdk, :process_module is still read as a fallback for query streaming, but it is deprecated and logs a warning once per legacy module.

SessionStore now hydrates on-disk cache in a handle_continue/2 step. Startup is faster, but list/search can be briefly incomplete immediately after boot while warmup finishes.

Transport.Port, Transport.Erlexec, and Streaming.Session support startup_mode: :lazy to defer subprocess startup to handle_continue/2. In lazy mode, start_link can succeed before the subprocess is spawned; startup failures then surface as process exit after init.

Query-side transport errors normalize equivalent reasons to stable atoms where possible: :port_closed is treated as :not_connected, and {:command_not_found, "claude"} is treated as :cli_not_found.

SDK Logging

The SDK uses its own log level filter (default: :warning) to keep output quiet in dev. Configure via application env:

config :claude_agent_sdk, log_level: :warning  # :debug | :info | :warning | :error | :off

Option Presets

alias ClaudeAgentSDK.OptionBuilder

# Environment-based presets
OptionBuilder.build_development_options()  # Permissive, verbose
OptionBuilder.build_production_options()   # Restrictive, safe
OptionBuilder.for_environment()            # Auto-detect from Mix.env()

# Use-case presets
OptionBuilder.build_analysis_options()     # Read-only code analysis
OptionBuilder.build_chat_options()         # Simple chat, no tools
OptionBuilder.quick()                      # Fast one-off queries

Examples

The examples/ directory contains runnable demonstrations.

Mix Task Example (Start Here)

If you want to integrate Claude into your own Mix project, see the mix_task_chat example — a complete working app with Mix tasks:

cd examples/mix_task_chat
mix deps.get
mix chat "Hello, Claude!"           # Streaming response
mix chat --interactive              # Multi-turn conversation
mix ask -q "What is 2+2?"           # Script-friendly output

Script Examples

# Run all examples
bash examples/run_all.sh

# Run a specific example
mix run examples/basic_example.exs
mix run examples/streaming_tools/quick_demo.exs
mix run examples/hooks/basic_bash_blocking.exs

Key Examples:

• mix_task_chat/ - Full Mix task integration (streaming + interactive chat)
• basic_example.exs - Minimal SDK usage
• streaming_tools/quick_demo.exs - Real-time streaming
• hooks/complete_workflow.exs - Full hooks integration
• sdk_mcp_tools_live.exs - Custom MCP tools
• advanced_features/agents_live.exs - Multi-agent workflows
• advanced_features/subagent_spawning_live.exs - Parallel subagent coordination
• advanced_features/web_tools_live.exs - WebSearch and WebFetch

Full Application Examples

Complete Mix applications demonstrating production-ready SDK integration patterns:

# Run Phoenix Chat
cd examples/phoenix_chat && mix deps.get && mix phx.server
# Visit http://localhost:4000

# Run Document Generation
cd examples/document_generation && mix deps.get && mix generate.demo

# Run Research Agent
cd examples/research_agent && mix deps.get && mix research "quantum computing"

# Run Skill Invocation demo
cd examples/skill_invocation && mix deps.get && mix run -e "SkillInvocation.demo()"

# Run Email Agent
cd examples/email_agent && mix deps.get && mix email.assistant "find emails from last week"

Guides

Upgrading

For breaking changes and migration notes, see CHANGELOG.md.

0.11.0 breaking changes:

• --print flag removed from all modules. All queries now use --output-format stream-json exclusively.
• --agents CLI flag removed. Agents are now sent via the initialize control request. Use Options.agents_for_initialize/1.
• AgentsFile module deleted. Remove any agents_temp_file_max_age_seconds config.
• Client state is now a defstruct. Four deprecated fields removed: current_model, pending_model_change, current_permission_mode, pending_inbound_count.
• All 12 hook events are now supported (6 new: post_tool_use_failure, notification, subagent_start, permission_request, session_start, session_end).

0.10.0 fix (resume turn persistence):

• resume/3 no longer uses --print --resume (one-shot mode that dropped intermediate turns). It now uses --resume with --input-format stream-json, preserving the full conversation history across resume calls.
• Updated default Opus model to claude-opus-4-6.

0.9.0 breaking change (streaming):

• Stream event wrappers now require uuid and session_id. Missing keys raise and terminate the streaming client.
• If you emit or mock stream_event wrappers, include both fields (custom transports, fixtures, tests).

Additional Resources:

• CHANGELOG.md - Version history and release notes
• HexDocs - API documentation
• Claude Code SDK - Upstream documentation

License

MIT License - see LICENSE for details.