sora-2-generate（索拉 2 生成）

Sora 2 Generate

Generate videos through the NanoPhoto.AI Sora 2 API.

Prerequisites

1. Obtain an API key at: https://nanophoto.ai/settings/apikeys
2. Configure NANOPHOTO_API_KEY before using the skill.
3. Do not paste the API key into chat; store it in the platform's secure env setting for this skill.

Preferred OpenClaw setup:

• Open the skill settings for this skill
• Add an environment variable named NANOPHOTO_API_KEY
• Paste the API key as its value

Equivalent config shape:

{
  "skills": {
    "entries": {
      "sora-2-generate": {
        "enabled": true,
        "env": {
          "NANOPHOTO_API_KEY": "your_api_key_here"
        }
      }
    }
  }
}

Other valid ways to provide the key:

• Shell: export NANOPHOTO_API_KEY="your_api_key_here"
• Tool-specific env config: any runtime that injects NANOPHOTO_API_KEY
• OpenClaw config fallback: the bundled script also falls back to ~/.openclaw/openclaw.json at skills.entries.sora-2-generate.env.NANOPHOTO_API_KEY

Credential declaration summary:

• Primary credential: NANOPHOTO_API_KEY
• Resolution order in the bundled script: --api-key → NANOPHOTO_API_KEY environment variable → ~/.openclaw/openclaw.json skill env
• No unrelated credentials are required

Choose the mode

• Use textToVideo when the user gives only a prompt.
• Use imageToVideo when the user provides one or more public image URLs.
• Do not use local image files or base64 images for the API. The API only accepts public imageUrls.

Recommended workflow

1. Collect the prompt.
2. Decide mode: textToVideo or imageToVideo.
3. Choose modelTier, aspectRatio, and videoDuration.

- Default to sora2. - Use sora2-pro-standard or sora2-pro-high only if the user explicitly wants the pro tier or higher resolution output.

1. Submit the generation request.
2. If the user wants synchronous progress output in the same process, use submit --follow.
3. Wait 120 seconds before the first status check, then check every 20 seconds until completed or failed.
4. Return the final videoUrl, generation time, and credits used when available.
5. Expect real-world wait time to vary from roughly 120 seconds to 480 seconds depending on queue/load, model tier, and prompt complexity; avoid short timeouts.

Preferred commands

Use the single bundled script with subcommands.

Submit only

python3 scripts/sora2_generate.py submit \
  --prompt "A golden retriever running on a beach at sunset, cinematic lighting" \
  --mode textToVideo \
  --model-tier sora2 \
  --aspect-ratio landscape \
  --video-duration 10

Submit and keep polling in the same process

python3 scripts/sora2_generate.py submit \
  --prompt "A golden retriever running on a beach at sunset, cinematic lighting" \
  --mode textToVideo \
  --model-tier sora2 \
  --aspect-ratio landscape \
  --video-duration 10 \
  --follow

Check status of an existing task

python3 scripts/sora2_generate.py status --task-id task_abc123

Image to video

python3 scripts/sora2_generate.py submit \
  --prompt "The person in the painting comes alive, moving naturally and reciting poetry" \
  --mode imageToVideo \
  --image-url https://static.nanophoto.ai/demo/nano-banana-pro.webp \
  --model-tier sora2 \
  --aspect-ratio landscape \
  --video-duration 10

Script behavior

The bundled script resolves credentials in this order: --api-key, then NANOPHOTO_API_KEY from the environment, then ~/.openclaw/openclaw.json at skills.entries.sora-2-generate.env.NANOPHOTO_API_KEY.

Subcommands:

• submit: submit a task
• submit --follow: submit and keep polling in the same process
• status: check an existing taskId

Cross-platform note:

• Use python3 on macOS/Linux.
• Use python on Windows unless python3 is available.
• The script uses Python's standard HTTP client and does not require curl.
• Use --json-only when another script/tool needs raw JSON output.
• Use --initial-status-delay to override the default 120-second wait before the first status check.
• Use --status-check-interval to override the default 20-second interval between progress checks.
• Default max wait is 1200 seconds.

Manual API calls

Submit generation

curl -X POST "https://nanophoto.ai/api/sora-2/generate" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NANOPHOTO_API_KEY" \
  --data-raw '{
    "prompt": "A golden retriever running on a beach at sunset, cinematic lighting",
    "mode": "textToVideo",
    "modelTier": "sora2",
    "aspectRatio": "landscape",
    "videoDuration": "10"
  }'

Check status

curl -X POST "https://nanophoto.ai/api/sora-2/check-status" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NANOPHOTO_API_KEY" \
  --data-raw '{"taskId": "task_abc123"}'

Parameter guidance

• modelTier

- sora2: default; lowest cost and fastest turnaround - sora2-pro-standard: 720p pro tier; use only when explicitly requested - sora2-pro-high: 1080p pro tier; use only when explicitly requested

• aspectRatio

- portrait: vertical/default - landscape: horizontal

• videoDuration

- 10 - 15

Error handling

Bundled files

• scripts/sora2_generate.py: single entry point for submit, status, and optional in-process polling.
• references/api.md: condensed API reference, costs, and response shapes.