Edit an image

You're Claude Code, working in my repository.

Task
Add an edit_image function to my app that sends POST /v1/images/edits with muse-image-1.0, a prompt and one or more input images (multipart through the OpenAI SDK, or Meta's JSON images array over HTTP) and saves data[].b64_json.

Read this first
Meta's docs page "Edit an image" as Markdown: https://dev.meta.ai/docs/api-reference/images/edit-image.md
It's the source of truth for request and response shapes. The same page for people: https://dev.meta.ai/docs/api-reference/images/edit-image

Setup
- Base URL: https://api.meta.ai/v1
- Auth: send Authorization: Bearer $MODEL_API_KEY, reading MODEL_API_KEY from the environment. Never hard-code, print or commit the key.
- Model: muse-image-1.0
- Python: the OpenAI SDK (pip install openai) with base_url="https://api.meta.ai/v1" and api_key=os.environ["MODEL_API_KEY"] works, or plain requests.

Facts to get right
- Multipart (client.images.edit): image for one file, or image[0], image[1] for several (never both), and each scalar field once. mask, input_fidelity and partial_images are 400 unsupported_parameter.
- JSON (requests or curl): images is 1 to 10 items, each exactly one of image_url (a public URL or a data URL) or file_id. Unknown item members are 400s, and a fetched image_url must be image/*.
- Other fields as generations: n 1 to 10, size (aspect ratio), response_format, output_format, reasoning_strength, temperature, user, tool_enablement, stream (image_edit.completed events).
- The reply is an ImagesResponse, the same shape as generations. $0.01 per image.

Starter code (Python)
import base64
import os

import requests

with open("fox.png", "rb") as f:
    image = "data:image/png;base64," + base64.b64encode(f.read()).decode()

response = requests.post(
    "https://api.meta.ai/v1/images/edits",
    headers={"Authorization": f"Bearer {os.environ['MODEL_API_KEY']}", "Content-Type": "application/json"},
    json={
        "model": "muse-image-1.0",
        "prompt": "add a small red wool hat on the fox's head, keep the snowy forest background",
        "images": [{"image_url": image}],
        "n": 1,
    },
)
response.raise_for_status()
with open("edited.webp", "wb") as f:
    f.write(base64.b64decode(response.json()["data"][0]["b64_json"]))

Done when
- Editing a local PNG through the SDK writes a changed image.
- The JSON body with an image_url works over plain HTTP.
- Two input images compose into one scene.
- 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.