About this skill

What It Solves

When an agent needs to generate images from text, manually assembling the API request, storing the token, parsing JSON, and saving files can interrupt the main workflow. sensenova-u1-image condenses this into a reusable text-to-image flow: call SenseNova U1 with curl, extract dynamic image URLs, and place the files where the user can view them.

How It Works

The skill first checks MEMORY.md for an existing sensenova-u1-image token; if none is found, it asks for the Authorization: Bearer value and saves it. During normal use, it:
- Collects prompt, size, and n while preserving the user's original language;
- Builds a curl request for sensenova-u1-fast;
- Captures the JSON response and parses image URLs dynamically;
- Downloads the images locally via browser_use or curl;
- Shows them through view_image or send_file_to_user.

Boundaries

It depends on a valid API token and network access, and does not include prompt optimization, image editing, or batch job scheduling. If the API returns an error, the full message should be surfaced. Use only the listed aspect ratios for size; do not assume undocumented values.

Screenshots

Use Cases

In an agent workflow, generate images from copy, poster concepts, or UI sketch descriptions using SenseNova U1.
Convert an English product description into a 1:1 image and save the returned URL locally for preview.
Verify in automation that the image API returns JSON URLs, then download the first image with curl.
Produce cover images on request using documented aspect ratios and image counts, then show the results.

Best For

AI workflow developers who need agents to generate and save images from descriptions
Test engineers who need to validate SenseNova text-to-image responses and download flows
Operations or design collaborators who need quick images for copy, posters, or covers
Integration engineers who need to wire prompt, size, and count parameters into agent tools