Zyt text to digital person

Use Chanjing text-to-digital-person APIs to create AI portrait images, turn them into talking videos, optionally run LoRA training, poll async tasks, and exp...

Security Scan

The skill's code, instructions, and requested environment variables are consistent with its stated purpose (calling Chanjing AIGC APIs, polling async tasks, and optionally saving results); it stores/refreshes tokens in a local credentials file and does not exhibit unexplained network or system access.

Purpose & Capability

The name/description (text-to-digital-person, image->video, LoRA) matches the implemented API calls and scripts: POST/GET to open-api.chanjing.cc endpoints for photo, motion, and lora tasks. Declared env vars (config dir and auto-open flag) relate to local credential handling and login UI behavior.

Instruction Scope

SKILL.md and scripts only instruct the agent to call the documented Chanjing endpoints, poll for async task status, and download results only on explicit user request. The runtime instructions and scripts read/write a local credentials file (~/.chanjing/credentials.json) and may open a browser for login if configured; there are no instructions to read unrelated system files or send data to unexpected endpoints.

Install Mechanism

No install spec or external downloads are present; this is an instruction-only skill with bundled Python scripts. Nothing is fetched from third-party URLs or written to non-standard system locations by an installer.

Credentials

The skill requests CHANJING_CONFIG_DIR and CHANJING_AUTO_OPEN_LOGIN which are proportional to its local-config behavior. However, actual secret credentials (app_id/secret_key) are expected in a local file (~/.chanjing/credentials.json) rather than as declared environment variables or a primary credential field; the skill will persist access_token and expire_in into that file. Users should be aware secrets are stored on-disk and managed by the scripts.

Persistence & Privilege

always is false and the skill does not modify other skills or system-wide agent configuration. It does persist credentials and tokens under ~/.chanjing (or $CHANJING_CONFIG_DIR) and attempts to set restrictive file permissions when writing the config file.

Assessment

This skill appears to do what it claims: call Chanjing's API to generate images/videos and run LoRA tasks. Before installing, verify the API base (https://open-api.chanjing.cc) is the expected provider and that you trust it. Understand that credentials (app_id and secret_key) are stored in a local file (~/.chanjing/credentials.json) and the scripts will write an access_token and expiry there; if you prefer, keep credentials in a secure location and set CHANJING_CONFIG_DIR accordingly. CHANJING_AUTO_OPEN_LOGIN can make the skill open a browser login page automatically—keep it unset if you don't want that behavior. If you have concerns, inspect the two included Python scripts (_auth.py and _task_api.py) yourself; they perform the token management and HTTP calls and do not appear to contact any endpoints beyond the documented Chanjing API.

Chanjing Text To Digital Person

When to Use This Skill

当用户要做这些事时使用本 Skill：

• 根据人物提示词生成数字人形象图
• 把生成的人物图转成会说话的短视频
• 查询文生图 / 图生视频 / LoRA 任务状态
• 在用户明确要求时，把生成图片或视频下载到本地

如果需求是“上传真人素材训练定制数字人”，优先使用 chanjing-customised-person。
如果需求是“拿已有数字人做口播视频合成”，优先使用 chanjing-video-compose。

Preconditions

本 Skill 自己包含本地配置和鉴权流程，不依赖其他 skill 的运行时脚本。

本 Skill 使用：

• 配置文件：~/.chanjing/credentials.json
• 若设置环境变量 CHANJING_CONFIG_DIR：使用 $CHANJING_CONFIG_DIR/credentials.json
• API 基础地址固定：https://open-api.chanjing.cc

当本地缺少 AK/SK 或 AK/SK 无效时，脚本默认返回登录引导信息，不自动打开浏览器。
如需本地自动开页，可显式设置：CHANJING_AUTO_OPEN_LOGIN=1。 https://www.chanjing.cc/openapi/login

Standard Workflow

主流程通常分两段，且都是异步任务：

1. 调用 create_photo_task 创建文生图任务，得到 photo_unique_id
2. 调用 poll_photo_task 轮询到成功，选一张 photo_path
3. 调用 create_motion_task 创建图生视频任务，得到 motion_unique_id
4. 调用 poll_motion_task 轮询到成功，得到最终 video_url
5. 只有在用户明确要求保存到本地时，才调用 download_result

可选扩展：

• 若用户想做 LoRA 训练，调用 create_lora_task 和 poll_lora_task
• poll_lora_task 成功后会返回一条 photo_task_id，可继续用 poll_photo_task 拿图

Covered APIs

本 Skill 当前覆盖：

• POST /open/v1/aigc/photo
• GET /open/v1/aigc/photo/task
• GET /open/v1/aigc/photo/task/page
• POST /open/v1/aigc/motion
• GET /open/v1/aigc/motion/task
• POST /open/v1/aigc/lora/task/create
• GET /open/v1/aigc/lora/task

Scripts

脚本目录：

• scripts/

脚本	说明
chanjing-config	写入/查看本地 app_id 与 secret_key，并清理旧 token 缓存
chanjing-get-token	从本地凭证获取有效 access_token（必要时自动刷新）
_auth.py	读取凭证、获取或刷新 access_token
create_photo_task	创建文生图任务，输出 photo_unique_id
get_photo_task	获取单个文生图任务详情
list_tasks	列出文生图任务列表；返回中 type=1 为 photo，type=2 为 motion
poll_photo_task	轮询文生图任务直到完成，默认输出第一张图片地址
create_motion_task	创建图生视频任务，输出 motion_unique_id
get_motion_task	获取单个图生视频任务详情
poll_motion_task	轮询图生视频任务直到完成，默认输出视频地址
create_lora_task	创建 LoRA 训练任务，输出 lora_id
get_lora_task	获取 LoRA 任务详情
poll_lora_task	轮询 LoRA 任务直到完成，默认输出第一条 photo_task_id
download_result	下载图片或视频到 outputs/text-to-digital-person/

Usage Examples

示例 1：文生图后直接图生视频

PHOTO_TASK_ID=$(python3 scripts/create_photo_task \
  --age "Young adult" \
  --gender Female \
  --number-of-images 1 \
  --industry "教育培训" \
  --background "现代直播间背景" \
  --detail "短发，亲和力强，职业装" \
  --talking-pose "上半身特写，站立讲解")

PHOTO_URL=$(python3 scripts/poll_photo_task \
  --unique-id "$PHOTO_TASK_ID")

MOTION_TASK_ID=$(python3 scripts/create_motion_task \
  --photo-unique-id "$PHOTO_TASK_ID" \
  --photo-path "$PHOTO_URL" \
  --emotion "自然播报，语气清晰自信" \
  --gesture)

python3 scripts/poll_motion_task \
  --unique-id "$MOTION_TASK_ID"

示例 2：LoRA 训练

LORA_ID=$(python3 scripts/create_lora_task \
  --name "演示LoRA" \
  --photo-url https://example.com/1.jpg \
  --photo-url https://example.com/2.jpg \
  --photo-url https://example.com/3.jpg \
  --photo-url https://example.com/4.jpg \
  --photo-url https://example.com/5.jpg)

python3 scripts/poll_lora_task \
  --lora-id "$LORA_ID"

Download Rule

下载是显式动作，不是默认动作：

• poll_photo_task 和 poll_motion_task 成功后应先返回远端 URL
• 不要自动下载结果文件
• 只有当用户明确表达“下载到本地”“保存到 outputs”“帮我落盘”时，才执行 download_result

Output Convention

默认本地输出目录：

• outputs/text-to-digital-person/

Additional Resources

更多接口细节见：

• reference.md
• examples.md