generate_content
Generate social media post suggestions with AI-written captions, titles, hashtags, and platform-specific adaptations. Returns multiple text suggestions (title, body, tags). For IMAGE format, each suggestion includes imagePrompts — use generate_image to create visuals. For VIDEO format, each suggestion includes videoPrompt (single-scene) or videoScenes (multi-scene) — use generate_video with the video prompt. After the user picks a suggestion, generate the visual separately. Use list_businesses first to get a valid businessId.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | Describe the content topic, e.g. 'Morning yoga tips for beginners' or 'New product launch announcement for our coffee shop' | |
| language | No | Override the output language (e.g. 'he', 'en', 'es'). If omitted, content is generated in the language matching the business's Content DNA / posting history. Only set this when the user explicitly requests content in a different language than their usual posts. | |
| businessId | Yes | Business ID to generate for. Call list_businesses first to get this. | |
| contentFormat | No | Target media format: IMAGE (visual post with caption), VIDEO (short video/reel), CAROUSEL (multi-image), REEL (short-form video), STORY (ephemeral), TEXT (caption only). Defaults to IMAGE. | |
| useContentDNA | No | Analyze the business's past viral posts to match their proven tone and style. Defaults to false. | |
| suggestionCount | No | Number of content suggestions to generate (1-6). Defaults to 4. Use 2 for quick scheduling flows where the user already knows what they want. |