# Wegent Image Generation

> Use this skill when the user asks to generate, draw, create, or revise an image. Supports text-to-image and image-reference generation.

- Skill: `wecode-ai/wegent-image-generation` (Agent Skill)
- Install (CLI): `npx skillmds@latest add wecode-ai/wegent-image-generation`
- Raw SKILL.md: https://api.skillmd.com/api/skills/wecode-ai/wegent-image-generation/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: wecode-ai (https://skillmd.com/u/wecode-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/wecode-ai/wegent-image-generation

---


# Image Generation

Call `generate_image` when the user asks for an image.

- Put the complete visual description in `prompt`.
- Use `reference_images` only when the user supplied image attachments or URLs.
- Pass attachment IDs, public HTTP/HTTPS URLs, or base64 data URLs, not local
  sandbox paths.
- For uploaded or previously generated images, pass attachment IDs. Do not pass Wegent
  attachment download URLs or relative `/api/attachments/...` paths as remote references.
- Default to one image unless the user explicitly asks for multiple images.
- The tool saves generated images as task attachments and returns image blocks. Do not create a duplicate card.
- If the tool returns `error`, explain it once and do not retry unchanged parameters.