generate_image
Generate an AI image for a show (use this for any thumbnail / cover art / artwork / image request — NOT the writing/article path). Runs in the background and saves the image to the workspace media library, attached to the entity in context so it appears on that entity's media: pass attach_to_type + attach_to_id (show | episode | social | contact | group | interview) — e.g. an episode thumbnail attaches to that episode; defaults to the show. Aspect ratio is controlled ONLY by the aspect_ratio parameter (1:1 | 16:9 | 9:16 | 4:5) — never by words in the prompt; the output is guaranteed to be exactly that ratio. Pick it from intent (YouTube thumbnail → 16:9, Short/Reel → 9:16) without asking. Optionally pass reference_media_ids (med_<uuid> from list_entity_media / list_media_assets) to edit/composite against real workspace images (host/guest photos, a logo); with no references it is pure text-to-image and cannot retain a supplied image. Returns a job_id — the image appears when the job completes; then apply_media_asset can set it as a cover/social image. Required: show_id, prompt. Billed in credits.
Request
- Base URL: https://app.podmule.com/api/v1
- URL: https://app.podmule.com/api/v1/generate_image
- Auth: HTTP bearer
Headers
Stripe-style idempotency key. When present, repeat invocations within 24h with the same body return the original response; same key + different body returns 409 IDEMPOTENCY_KEY_REUSED.
When set to "true", routes the call to the capability's dryRun() method — same scope/precondition checks fire as the real handler, but persistence is skipped. The audit row is tagged result_status=dry_run.
Request body
Response
Capability output wrapped in { data }.
Changes
No recorded changes to this operation across all 1 revision of this API.