Skip to content
WP EngineDocumentation

Create a message (Anthropic Messages API)

POST
/v1/messages
<?php


$client = new \GuzzleHttp\Client();


$response = $client->request('POST', 'https://api.ai.wpengine.com/v1/messages', [
  'body' => '{ "model": "anthropic/claude-sonnet-4-6", "messages": [ { "role": "user", "content": "example" } ], "max_tokens": 1024, "system": "example", "tools": [ {} ], "stream": false, "thinking": { "type": "enabled", "budget_tokens": 1 } }',
  'headers' => [
    'Authorization' => 'Bearer <token>',
    'Content-Type' => 'application/json',
  ],
]);


echo $response->getBody();

Anthropic-native Messages endpoint, compatible with official Anthropic SDKs. Accepts the standard Anthropic Messages request shape, including system prompts, multi-turn conversations, image content blocks, tool use, extended thinking, custom stop sequences, and streaming. Prompt caching is supported: add “cache_control” breakpoints to your request (on system, tools, or message content) to cache stable prefixes; cached-token counts are reported as “cache_creation_input_tokens” and “cache_read_input_tokens” in the response usage, and “cache_creation” splits the cache writes by “cache_control” ttl. A 1-hour cache write is charged at a higher rate than a 5-minute one. Authenticate with a WP Engine bearer token, or with the Anthropic SDK convention of an “x-api-key” header carrying a WP Engine API key. API keys must carry the “chat:write” permission. The model field accepts prefixed (“anthropic/claude-sonnet-4-6”) or bare (“claude-sonnet-4-6”) names; only Anthropic models are supported. The “anthropic-version” and “anthropic-beta” headers are accepted and forwarded. All errors are returned in Anthropic’s error envelope format.

Authorizations

Section titled “Authorizations”

Request Bodyrequired

Section titled “Request Bodyrequired”
Media typeapplication/json

Standard Anthropic Messages API request. The full request shape is supported and forwarded to the model, including fields not listed here (tool_choice, stop_sequences, temperature, top_p, metadata, …). See Anthropic’s Messages API reference for the complete schema.

object
model
required

Anthropic model name, prefixed or bare (e.g. “anthropic/claude-sonnet-4-6” or “claude-sonnet-4-6”).

string
Example
anthropic/claude-sonnet-4-6
messages
required

Conversation turns in Anthropic message format.

Array<object>
object
role
required
string
Allowed values: user assistant
content
required

Message content: a plain string, or an array of content blocks (text, image, tool_use, tool_result, thinking, …).

max_tokens
required

Maximum number of tokens to generate.

integer
>= 1
Example
1024
system

System prompt, as a string or array of content blocks.

tools

Tool definitions in Anthropic tool format, each with a name, description, and input_schema.

Array<object>
object
stream

When true, the response is a Server-Sent Events stream using Anthropic’s streaming event protocol.

boolean
thinking

Extended thinking configuration.

object
type
string
Allowed values: enabled disabled
budget_tokens

Token budget for thinking when enabled.

integer

Responses

Section titled “Responses”

200

Section titled “200”

Anthropic message response. When stream is false, returns a single JSON message object. When stream is true, returns Server-Sent Events using Anthropic’s streaming event protocol (message_start, content_block_delta, message_delta, message_stop, …).

Standard Anthropic Messages API response.

object
id
string
type
string
Allowed values: message
role
string
Allowed values: assistant
model
string
content

Generated content blocks (text, tool_use, thinking, …).

Array<object>
object
type
string
stop_reason
string
Allowed values: end_turn max_tokens stop_sequence tool_use model_context_window_exceeded
stop_sequence

The matched custom stop sequence, when applicable.

string
nullable
usage
object
input_tokens
integer
output_tokens
integer
cache_creation_input_tokens
integer
cache_read_input_tokens
integer
cache_creation

Breakdown of cache_creation_input_tokens by the cache_control ttl the prompt cache was written with. A 1-hour write is charged at a higher rate than a 5-minute one.

object
ephemeral_5m_input_tokens

Tokens written to the cache with a 5-minute ttl.

integer
ephemeral_1h_input_tokens

Tokens written to the cache with a 1-hour ttl.

integer
output_tokens_details

Breakdown of output_tokens. On streams it appears only on the final message_delta.

object
thinking_tokens

Thinking tokens, already counted in output_tokens.

integer
server_tool_use

Server-side tool calls the provider ran, each charged in credits on top of tokens. On streams the count is cumulative on message_delta.

object
web_search_requests

Web searches run for this message.

integer
Example
{
  "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
  "type": "message",
  "role": "assistant",
  "model": "claude-sonnet-4-6",
  "content": [
    {
      "type": "text"
    }
  ],
  "stop_reason": "end_turn"
}

400

Section titled “400”

Invalid request body or parameters, in Anthropic error envelope format (error type “invalid_request_error”).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

401

Section titled “401”

Missing or invalid credentials, in Anthropic error envelope format (error type “authentication_error”).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

Headers

Section titled “Headers”
WWW-Authenticate
string
Example
Bearer realm="ai-services"

402

Section titled “402”

The account’s credit balance is exhausted. Grant or top up credits before retrying. Body is in Anthropic error envelope format.

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

403

Section titled “403”

Valid credentials but insufficient permission, in Anthropic error envelope format (error type “permission_error”). The same type is returned when billing isn’t set up for the account yet, because AI hasn’t been turned on for it in the User Portal or its setup hasn’t finished; the message says so. On /v1/messages, when content moderation is enabled, a 403 with error type “content_policy_violation” instead means the request content was blocked by content moderation policy; that response never identifies which filter or category matched.

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

404

Section titled “404”

The requested model was not found or does not support this endpoint, in Anthropic error envelope format (error type “not_found_error”).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

413

Section titled “413”

Request body exceeds the size limit, in Anthropic error envelope format (error type “request_too_large”).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

429

Section titled “429”

Rate limit exceeded. The request was throttled at the edge; wait and retry later. The response body is a short HTML page generated by the load balancer, not a structured JSON error — clients should rely on the 429 status code rather than parsing the body.

Media typetext/html
string

502

Section titled “502”

Upstream provider returned an error, in Anthropic error envelope format (error type “api_error”).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

503

Section titled “503”

A required dependency (such as the authorization service) is temporarily unavailable, in Anthropic error envelope format (error type “overloaded_error”). Retry after a short delay with exponential back-off.

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}

default

Section titled “default”

Upstream provider error relayed with its original status code and translated into Anthropic error envelope format. The error type reflects the status code (e.g. “rate_limit_error” for 429, “overloaded_error” for 529).

Media typeapplication/json

Anthropic error envelope, as parsed by Anthropic SDKs.

object
type
required
string
Allowed values: error
error
required
object
type
required

Anthropic error type, e.g. “invalid_request_error”, “authentication_error”, “permission_error”, “not_found_error”, “request_too_large”, “rate_limit_error”, “overloaded_error”, “api_error”, or “content_policy_violation” (request content blocked by moderation policy when content moderation is enabled — never identifies which filter or category matched).

string
message
required
string
request_id

Correlation ID for this request, matching the X-Request-ID response header. Quote it in support reports.

string
Example
{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "max_tokens is required and must be greater than zero",
    "request_id": "80f1e2c3a4b5c6d7e8f9a0b1c2d3e4f5"
  }
}