Aller au contenu principal
Ce contenu n'est pas encore disponible dans votre langue et est affiché en anglais.

MeiGen AI Design

Plugin Vérifié
92

AI image generation with creative workflow orchestration, prompt engineering, and curated inspiration library via MCP server

Résumé IA

This plugin enables AI image generation through its integrated agents and commands, supporting MeiGen Cloud, local ComfyUI, and OpenAI-compatible APIs. It provides features for prompt crafting, parallel generation, and searching a large inspiration gallery, with clear setup instructions for various providers.

Documentation

  • info:Configuration & parameter referenceThe README documents configuration options for providers (API tokens, URLs) and environment variables, but specific default values for image generation parameters like aspect ratio or model selection are not explicitly detailed, only inferred from tool descriptions.

Maintenance

  • warning:Dependency ManagementThe plugin lists dependencies like `meigen@1.2.13` in `.mcp.json` and `Node.js 18+` in its README, but there is no explicit mention of vulnerability scanning or automated update mechanisms for these dependencies.

Security

  • warning:Secret ManagementAPI keys and tokens are expected to be set as environment variables (e.g., `MEIGEN_API_TOKEN`, `OPENAI_API_KEY`). While this is a common practice, there's no explicit mention of how these secrets are protected or if they are ever exposed in logs or debug output.
  • warning:Keychain-stored secretsSecrets like API keys are handled via environment variables. The plugin.json indicates a MIT license, but there's no explicit mechanism described for routing secrets through `userConfig` with `sensitive: true` for keychain storage, potentially leading to secrets being stored in plain text settings.

Code Execution

  • info:ValidationThe plugin uses agents and commands that interact with external services and user prompts. While prompts are enhanced, explicit schema validation for all inputs and outputs (beyond what the MCP server might provide) is not detailed.
  • info:Error HandlingThe agent definitions indicate delegation to specific tools and models. While error handling is implied by the structure, explicit details on how errors are categorized, reported meaningfully, or handled by the agent beyond tool response are not provided.

Installation

Ajoutez d'abord la marketplace

/plugin marketplace add wshobson/agents
/plugin install meigen-ai-design@claude-code-workflows
Mis à jour le 4 days ago
Voir le code source