Integrations

ChatGPT

SteadyShot exposes a public OpenAPI spec, which is exactly what ChatGPT custom GPTs consume as their Actions. Build a GPT that can call any SteadyShot endpoint by natural language in under 10 minutes.

Build the GPT

  1. In ChatGPT, click your profile → My GPTs Create a GPT.
  2. In the Configure tab, scroll to Actions → click Create new action.
  3. Under Schema, click Import from URL and paste:
text
https://steadyshot.ai/openapi.json
  1. Under Authentication, choose API Key Bearer. Paste an API key from /api-keys.
  2. Add this to the GPT’s instructions so it picks tools correctly:
text
You can call SteadyShot to generate consistent character
illustrations. When the user asks to "make an image of [character]
doing [thing]", call list_characters first if you don't know the
character_id, then generate_scene with the matched id and the user's
prompt. For variations of an existing scene, use vary_scene with the
returned image_id. Storybooks: generate_storybook with an array of
page prompts.

Try it

  • “Show me my SteadyShot characters” → GPT calls list_characters.
  • “Generate Luna reading by a campfire” → GPT looks up Luna’s id, then calls generate_scene. Inline image preview.
  • “Now make 4 versions where she’s frowning” → GPT calls vary_scene with the last image_id.
Custom GPTs are gated behind a ChatGPT Plus / Team / Enterprise subscription on the OpenAI side. SteadyShot doesn’t require any extra plan — every action still bills against your SteadyShot credits.