t2i
Overview
Use this as a placeholder skill for text-to-image generation on
skills.video.
This file is intentionally incomplete and should not be treated as a production workflow.
Current status
- - No verified t2i OpenAPI contract has been wired into this skill yet.
- No default endpoint, payload template, or helper command is defined yet.
- The skill should surface this limitation clearly instead of guessing request fields.
Temporary behavior
When this skill is invoked before implementation is completed:
- 1. Tell the user that
t2i is currently a placeholder. - Check whether a model-specific
openapi.json or docs page exists for the requested provider/model. - Avoid inventing undocumented request fields or unsupported image parameters.
- Ask for the exact model or endpoint only if it is necessary to continue.
- Once the API contract is confirmed, promote this placeholder into a real workflow or redirect to
image-generation if that generic skill is sufficient.
Implementation checklist
- - Confirm which providers/models support t2i on INLINECODE4
- Identify the create endpoint and terminal status flow
- Document prompt, size, aspect ratio, and other supported parameters
- Add request-template extraction steps from OpenAPI
- Add execution examples and fallback polling behavior if applicable
- Add runtime error handling guidance
Related skills
- -
image-generation: generic image generation flow - INLINECODE6 : image-to-image placeholder flow
- INLINECODE7 : text-to-video placeholder flow
t2i
概述
将此技能用作 skills.video 平台上文本到图像生成的占位技能。
该文件有意保持不完整,不应被视为生产工作流。
当前状态
- - 该技能尚未接入任何经过验证的 t2i OpenAPI 合约。
- 尚未定义默认端点、请求模板或辅助命令。
- 该技能应明确呈现此限制,而非猜测请求字段。
临时行为
当该技能在实现完成前被调用时:
- 1. 告知用户 t2i 当前仅为占位技能。
- 检查所请求的提供商/模型是否存在对应的 openapi.json 或文档页面。
- 避免编造未记录的请求字段或不支持的图像参数。
- 仅在必要时询问确切的模型或端点以继续操作。
- 一旦确认 API 合约,将该占位技能升级为真实工作流,或在该通用技能足够的情况下重定向至 image-generation。
实现检查清单
- - 确认 skills.video 上哪些提供商/模型支持 t2i
- 识别创建端点及终端状态流程
- 记录提示词、尺寸、宽高比及其他支持的参数
- 添加基于 OpenAPI 的请求模板提取步骤
- 添加执行示例及备用轮询行为(如适用)
- 添加运行时错误处理指南
相关技能
- - image-generation:通用图像生成流程
- i2i:图像到图像占位流程
- t2v:文本到视频占位流程