foxapi-imagegen-skill
v0.1.3
Published
Install and configure the FoxAPI image generation Codex skill.
Readme
FoxAPI Imagegen Skill CLI
This package installs and configures the FoxAPI image generation skill for Codex.
One-time setup
For a new user, run this command locally:
npx --yes foxapi-imagegen-skill setupThe command installs the skill, prompts for a FoxAPI API key with hidden input, and saves the key in the user's Codex configuration directory. The key is not written into the Skill files or the npm package. Restart Codex after setup.
To replace an existing installation and update its saved key:
npx --yes foxapi-imagegen-skill setup --forceInstall from npm
npx --yes foxapi-imagegen-skill installThe shortest hosted command is:
npx --yes https://course.foxapi.cn/f.tgzOn Windows, the default destination is %USERPROFILE%\\.codex\\skills\\foxapi-imagegen.
On macOS and Linux, it is ~/.codex/skills/foxapi-imagegen.
To replace an existing copy:
npx --yes foxapi-imagegen-skill install --forceThe package does not contain API keys or generated images. The installed skill reads
the setup file, FOXAPI_API_KEY, or another supported credential source when it runs.
To configure a key without the setup command, set FOXAPI_API_KEY in the user's
environment. The setup command is preferred because it does not put the key in a
shell command history.
Run the installed skill
python ~/.codex/skills/foxapi-imagegen/scripts/foxapi_imagegen.py generate --prompt "A clean product illustration"Choose the new FoxAPI image models explicitly with their model IDs:
python ~/.codex/skills/foxapi-imagegen/scripts/foxapi_imagegen.py generate --model "gpt-image-2.5-flare" --prompt "A clean product illustration"
python ~/.codex/skills/foxapi-imagegen/scripts/foxapi_imagegen.py generate --model "gpt-image-2.5-sunburst" --prompt "A clean product illustration"Both models accept --size, including custom dimensions. The API may still return a different actual size or aspect ratio; use the reported actual_size value.
The skill calls https://foxapi.cn/v1/responses with the
image_generation tool and saves the returned Base64 image locally.
