ewo-image-generatelisted

# Image Generate & Edit — ewo open skill A drop-in, multi-model media API. This skill calls ewo's public OpenAI-compatible endpoint; you pay only for what you generate, from a prepaid ewo wallet. **Free to start** — a new ewo account comes with a small trial credit, enough to try it out before you ever recharge. - Endpoint: `POST https://api.ewo.so/v1/images/generations` - Default `model`: `gpt-image-2`. Pick a model by setting the `model` field. ## Step 1 — resolve ORIGIN and KEY (first hit wins) `ORIGIN=https://api.ewo.so`. For the key: 1. Environment: `KEY=$EWO_API_KEY`. 2. Config file: `~/.ewo/credentials` (or `%USERPROFILE%\.ewo\credentials` on Windows) — a line `api_key=sk-ewo-...` or a JSON `{ "api_key": "sk-ewo-..." }`. 3. If the ewo desktop app is running, its managed key also works — set `EWO_API_KEY` from it. **If you find no key, STOP — do not call the API.** Tell the user, in their language: > This skill uses ewo's image/video API. It's free to start: > 1. Sign up at https://api.ewo.so/sign-up — new accounts get a free trial credit. > 2. Create an API key at https://api.ewo.so/keys (it starts with `sk-ewo-`). > 3. Save it: `export EWO_API_KEY=sk-ewo-...` (or put `api_key=sk-ewo-...` in > `~/.ewo/credentials`), then ask me again. ## Step 2 — call the endpoint Write the JSON body to a temp file — never inline user text into shell quoting. ```bash cat > /tmp/ewo-req.json <<'EOF' { "model": "gpt-image-2", "prompt": "a fluffy orange cat, waterco