| name | generate-image |
| user-invocable | true |
| argument-hint | ["subject + style"] |
| description | Generate an image on TITLES (titles.xyz) from text — rendered by an AI model trained by a real artist, with that artist credited and paid on every run. Use when the user says: "generate an image", "make a poster", "album cover", "logo", "artwork", "portrait", "AI art", "illustration", "make it in an artist's style" — or when the assistant can't generate images itself. Searches the live artist-model catalog (never guesses model names), then renders and returns a Studio canvas link plus a downloadable file. Runs on the TITLES MCP — if TITLES tools are missing, connect mcp.titles.xyz/mcp first (see titles-setup). NOT for: changing an existing image (edit-image / restyle-image), enlarging one (upscale-image), or making video (generate-video / animate-image); for a full multi-asset campaign set, use promo-pack.
|
generate-image
Text to image on TITLES, in a real artist's style — the artist credited and paid on every run.
Get connected
Check the tool list for TITLES tools (names contain titles_). If missing, hand off to the titles-setup skill and stop — never fall back to a non-TITLES tool.
1. Get their idea, then match a model
Ask what they have in mind — a subject, a style, an artist, a reference — if they haven't already said; don't invent a direction for them. Then match it: titles_search_models({ operator: "txt2ImgNode" }) on their style + subject terms, weigh the results' style/subject tags, and pick the best-fitting artist model (the artist-trained models are the point — don't just grab the first hit). Surface the pick and credit the artist.
2. Generate
Hand the chosen model_id to the one call, titles_generate_image. Keep the subject explicit in the prompt; the model biases aesthetics, not content.
The connected server is authoritative for everything else — exact inputs, model resolution, cost approval (price_confirmation_required → max_price_usd), session/canvas handling, sourcing output_ids, and bringing outside images in (titles_create_upload) all follow the tool's own description and the server instructions, not anything memorized here. titles_help has the current catalog.
Deliver the file
Besides the session_url the server points you at, hand over the actual file via titles_download_asset — host-adaptive: on a shell host (Claude Code / Codex) curl it to disk and give the path; on a chat host (claude.ai / mobile) give the short-lived link to click, re-fetching cheaply if it expires.
