AI Media Generation

Generate images & videos with AIsa. Gemini 3 Pro Image (image) + Qwen Wan 2.6 (video) via one API key.

Security Scan

The skill's requirements, instructions, and included script are coherent with its stated purpose (generating images and videos via AIsa) and request only the single API key they need.

Purpose & Capability

Name/description (image + video generation via AIsa) matches the requested binary dependencies (python3, curl), the single environment variable (AISA_API_KEY), and the code which calls api.aisa.one endpoints. Nothing requested appears unrelated to media generation.

Instruction Scope

SKILL.md instructs use of curl and the provided Python client to call AIsa endpoints, create video tasks, poll status, and save results. The instructions do not attempt to read unrelated local files or extra environment variables. Note: the client will download external URLs returned by the API (signed media URLs) and will accept user-supplied img_url values, which means it may read or fetch arbitrary external hosts as part of normal operation — this is expected for a media downloader but worth user awareness.

Install Mechanism

There is no install spec; this is instruction-only plus a Python script. No packages or downloads are installed by the skill itself.

Credentials

Only a single API credential (AISA_API_KEY) is required and declared as the primary credential. The script reads that env var or accepts an explicit CLI argument. No other secrets or unrelated credentials are requested.

Persistence & Privilege

Skill does not request permanent inclusion (always is false), does not modify other skills or system-wide settings, and does not ask to store tokens in unrelated locations. It runs as an on-demand client script.

Assessment

This skill appears to be what it claims: a client for AIsa image/video generation that needs only your AISA_API_KEY. Before installing or running it: 1) Confirm you trust the AIsa service and that the API key you provide is intended for this use (don’t reuse high-privilege or unrelated keys). 2) Review the included script if you have concerns — it makes HTTPS calls to api.aisa.one, decodes inline base64 images, and writes files to disk. 3) Be mindful when supplying img_url values or allowing automatic downloads: the client will fetch/download URLs returned by the API (and will accept user-provided image URLs), so avoid passing local/internal URLs or sensitive endpoints. 4) Run it in a controlled environment first (not on sensitive hosts) and inspect outputs/paths it writes to. If you don’t trust the upstream API (api.aisa.one) or the key, don’t install/use the skill.

OpenClaw Media Gen 🎬

用 AIsa API 一把钥匙生成图片与视频：

• 图片：gemini-3-pro-image-preview（Gemini GenerateContent）
• 视频：wan2.6-t2v（通义万相 / Qwen Wan 2.6，异步任务）

API 文档索引见 AIsa API Reference（可从 https://docs.aisa.one/llms.txt 找到所有页面）。

🔥 你可以做什么

图片生成（Gemini）

"生成一张赛博朋克风格的城市夜景，霓虹灯，雨夜，电影感"

视频生成（Wan 2.6）

"用一张参考图生成 5 秒镜头：镜头缓慢推进，风吹动头发，电影感，浅景深"

Quick Start

export AISA_API_KEY="your-key"

---

🖼️ Image Generation (Gemini)

Endpoint

• Base URL: https://api.aisa.one/v1
• POST /models/{model}:generateContent

文档：google-gemini-chat（GenerateContent）见 https://docs.aisa.one/reference/generatecontent。

curl 示例（返回 inline_data 时为图片）

curl -X POST "https://api.aisa.one/v1/models/gemini-3-pro-image-preview:generateContent" \
  -H "Authorization: Bearer $AISA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents":[
      {"role":"user","parts":[{"text":"A cute red panda, ultra-detailed, cinematic lighting"}]}
    ]
  }'

说明：该接口的响应中可能出现 candidates[].parts[].inline_data（通常包含 base64 数据与 mime 类型）；客户端脚本会自动解析并保存文件。

---

🎞️ Video Generation (Qwen Wan 2.6 / Tongyi Wanxiang)

Create task

• Base URL: https://api.aisa.one/apis/v1
• POST /services/aigc/video-generation/video-synthesis
• Header：X-DashScope-Async: enable（必填，异步）

文档：video-generation 见 https://docs.aisa.one/reference/post_services-aigc-video-generation-video-synthesis。

curl -X POST "https://api.aisa.one/apis/v1/services/aigc/video-generation/video-synthesis" \
  -H "Authorization: Bearer $AISA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-DashScope-Async: enable" \
  -d '{
    "model":"wan2.6-t2v",
    "input":{
      "prompt":"cinematic close-up, slow push-in, shallow depth of field",
      "img_url":"https://upload.wikimedia.org/wikipedia/commons/thumb/3/3a/Cat03.jpg/320px-Cat03.jpg"
    },
    "parameters":{
      "resolution":"720P",
      "duration":5,
      "shot_type":"single",
      "watermark":false
    }
  }'

Poll task

• GET /services/aigc/tasks?task_id=...

文档：task 见 https://docs.aisa.one/reference/get_services-aigc-tasks。

curl "https://api.aisa.one/apis/v1/services/aigc/tasks?task_id=YOUR_TASK_ID" \
  -H "Authorization: Bearer $AISA_API_KEY"

---

Python Client

# 生成图片（保存到本地文件）
python3 {baseDir}/scripts/media_gen_client.py image \
  --prompt "A cute red panda, cinematic lighting" \
  --out "out.png"

# 创建视频任务（需要 img_url）
python3 {baseDir}/scripts/media_gen_client.py video-create \
  --prompt "cinematic close-up, slow push-in" \
  --img-url "https://upload.wikimedia.org/wikipedia/commons/thumb/3/3a/Cat03.jpg/320px-Cat03.jpg" \
  --duration 5

# 轮询任务状态
python3 {baseDir}/scripts/media_gen_client.py video-status --task-id YOUR_TASK_ID

# 等待直到成功（可选：成功后打印 video_url）
python3 {baseDir}/scripts/media_gen_client.py video-wait --task-id YOUR_TASK_ID --poll 10 --timeout 600

# 等待直到成功并自动下载 mp4
python3 {baseDir}/scripts/media_gen_client.py video-wait --task-id YOUR_TASK_ID --download --out out.mp4