Create a message

You're Claude Code, working in my repository.

Task
Add an Anthropic-format message call to my app against Meta Model API: model, max_tokens and messages, with system, custom tools, thinking and output_config.format as needed. Handle stop_reason tool_use by running the tool and sending tool_result in the next user turn, and replay the assistant content as returned.

Read this first
Meta's docs page "Create a message" as Markdown: https://dev.meta.ai/docs/api-reference/messages/create-message.md
It's the source of truth for request and response shapes. The same page for people: https://dev.meta.ai/docs/api-reference/messages/create-message

Setup
- Base URL: https://api.meta.ai/v1 for HTTP calls (POST https://api.meta.ai/v1/messages); the Anthropic SDK takes https://api.meta.ai
- Auth: send Authorization: Bearer $MODEL_API_KEY, reading MODEL_API_KEY from the environment. Never hard-code, print or commit the key.
- Model: muse-spark-1.3-contributor (Contributor mode: Meta may use what you send to train future models; drop -contributor for Standard mode)
- Python: the Anthropic SDK (pip install anthropic) with base_url="https://api.meta.ai" (no /v1) and auth_token=os.environ["MODEL_API_KEY"], or plain requests to https://api.meta.ai/v1/messages.

Facts to get right
- POST https://api.meta.ai/v1/messages; body AnthropicMessageRequest with model, messages (at least one) and max_tokens (at least 1).
- Custom tools carry input_schema; tool_choice takes auto, any or none (a named tool is a 400); disable_parallel_tool_use is accepted.
- content blocks: text, thinking (a summary, per the page) and redacted_thinking (encrypted; the only kind Meta sent when this was built): replay both. Also tool_use, server_tool_use (web_search). stop_reason: end_turn, tool_use, max_tokens or refusal.
- output_config.format {"type": "json_schema", "schema"} for structured output; output_config.effort low, medium, high, xhigh.
- stop_sequences, top_k, container and inference_geo are 400s; temperature runs 0 to 1.

Starter code (Python)
import json

import os

from anthropic import Anthropic

client = Anthropic(base_url="https://api.meta.ai", auth_token=os.environ["MODEL_API_KEY"])
MODEL = "muse-spark-1.3-contributor"
tools = [{"name": "get_weather", "description": "Get the current weather for a city.",
          "input_schema": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}]
messages = [{"role": "user", "content": "What's the weather in Paris right now?"}]

reply = client.messages.create(model=MODEL, max_tokens=1024, messages=messages, tools=tools)
if reply.stop_reason == "tool_use":
    messages.append({"role": "assistant", "content": reply.content})
    results = [{"type": "tool_result", "tool_use_id": b.id, "content": json.dumps({"temperature_c": 18})}
               for b in reply.content if b.type == "tool_use"]
    messages.append({"role": "user", "content": results})
    reply = client.messages.create(model=MODEL, max_tokens=1024, messages=messages, tools=tools)
print(reply.stop_reason, "".join(b.text for b in reply.content if b.type == "text"))

Done when
- A tool_use reply is answered with tool_result and the final reply ends with end_turn.
- A 400 (try top_k) surfaces the error envelope's message.
- The key comes from MODEL_API_KEY and never appears in code, logs or commits.
- API errors (4xx, 429) surface their message instead of failing silently.