grok-image-api
Work with OpenAI-compatible image generation and image editing endpoints. Use when the user wants to generate images from prompts, edit images with prompts and optional masks, test an image endpoint, or integrate /v1/images/generations or /v1/images/edits into scripts or projects.
- License 许可
- MIT
- 最近更新
- 2026-06-10
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- grok-image-api
- description
- Work with OpenAI-compatible image generation and image editing endpoints. Use when the user wants to generate images from prompts, edit images with prompts and optional masks, test an image endpoint, or integrate /v1/images/generations or /v1/images/edits into scripts or projects.
03 SKILL.md 正文
Grok Image API
Use this skill to work with image endpoints that expose:
POST /v1/images/generationsPOST /v1/images/edits
Prefer practical workflows: generate or edit an image, save the result locally, and then send or reuse the saved file.
Quick start
Prefer the bundled script:
scripts/grok_image_api.py generatefor text-to-imagescripts/grok_image_api.py editfor image editingscripts/grok_image_api.py probeto verify the endpoint is reachable
Default environment variables:
IMAGE_API_BASE_URLIMAGE_API_KEY
Workflow
- Identify the task: generate or edit.
- Load credentials through environment variables.
- For generation, send JSON to
/v1/images/generations. - For editing, send multipart form data to
/v1/images/editswithimageand optionalmask. - Save the returned image locally. If the API returns a URL, download it locally by default.
- If the user is on QQ and wants the image delivered, reply with a
<qqimg>tag pointing to the local file or returned URL.
Generation
Example:
IMAGE_API_BASE_URL="https://example.com/v1" \
IMAGE_API_KEY="..." \
python3 scripts/grok_image_api.py generate \
--style anime \
--preset-size portrait \
--prompt "a cozy cyberpunk tea shop at night" \
--out /tmp/generated.png
Use --size or --preset-size, --style, --model, --n, and --extra key=value when needed.
Editing
Example:
IMAGE_API_BASE_URL="https://example.com/v1" \
IMAGE_API_KEY="..." \
python3 scripts/grok_image_api.py edit \
--image /absolute/path/input.png \
--style cute \
--prompt "replace the background with a sunset beach" \
--out /tmp/edited.png
Add --mask /absolute/path/mask.png if the API supports masked edits.
Notes
- Assume this service is not a full chat-completions endpoint.
- Do not call unrelated routes unless the user explicitly says they exist.
- Avoid echoing secrets back to the user.
- Prefer local saved files for messaging channels that can upload local images.
- Default output directory is a local
output/grok-images/folder when--outis omitted.
References
Read references/api-notes.md when you need a compact reminder of request/response patterns, presets, and output behavior.
04 同仓库更多 Skills
youtube-content
YouTube transcripts to summaries, threads, blogs.
xurl
X/Twitter via xurl CLI: post, search, DM, media, v2 API.
touchdesigner-mcp
Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.
test-driven-development
TDD: enforce RED-GREEN-REFACTOR, tests before code.
teams-meeting-pipeline
Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.