panda-imagine

多 Provider 图片生成引擎。支持 OpenAI、Azure、Google Gemini、OpenRouter、 通义万相(DashScope)、MiniMax、即梦(Jimeng)、豆包(Seedream)、Replicate。 支持文生图、参考图、宽高比、批量生成。 触发词："生成图片"、"画图"、"g...

This skill appears to do what it says: unify multiple image-generation providers. Before installing, review and accept these points: (1) It will load/save preferences (EXTEND.md) and read .panda-skills/.env and ~/.panda-skills/.env (so API keys or other env values stored there can be read). (2) Runtime runs the included TypeScript via bun or npx -y bun—npx may fetch packages on demand. (3) For the Jimeng provider the skill can detect and run a local 'dreamina' CLI and suggests installing it via curl | bash (optional but be cautious fetching remote install scripts). (4) Only provide API keys with least privilege to providers you trust and avoid running the skill in environments containing unrelated high-value credentials. If you want stronger assurance, review the included scripts (providers/*.ts and main.ts) yourself or run the tool in an isolated environment/container.

Type: OpenClaw Skill Name: panda-imagine Version: 1.0.1 The panda-imagine skill bundle is a comprehensive multi-provider image generation tool supporting services like OpenAI, Google Gemini, DashScope, and others. The implementation in scripts/main.ts and the providers/ directory follows standard patterns for API integration, including environment variable management, configuration loading via EXTEND.md, and robust batch processing with rate limiting. While SKILL.md mentions a curl|bash command for the optional dreamina CLI, this is a standard installation method for that specific third-party tool and is not executed automatically. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.