trojanbox/mcp-feedback-elicitation
If you are the rightful owner of mcp-feedback-elicitation and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
An interactive feedback elicitation server based on the Model Context Protocol (MCP).
MCP Feedback Elicitation Server
An interactive feedback elicitation server based on the Model Context Protocol (MCP). It uses the MCP Elicitation feature to replace traditional Web UI solutions, providing a standardized user feedback collection experience.
Project Background
- š¤ AI Generated: This project was generated by Claude Sonnet 4.
- š Implementation Reference: Implemented with reference to mcp-feedback-enhanced. For more powerful features, such as image uploads, session management, etc., please use the mcp-feedback-enhanced project.
- š§ Technical Feature: Uses MCP Elicitation instead of a Web UI, requiring client support for the Elicitation feature.
Compatibility
Currently, VS Code 1.102+ supports the MCP Elicitation feature.
Quick Start
Add the following configuration to your mcp.json
:
"mcp-feedback-elicitation": {
"command": "npx",
"args": [
"mcp-feedback-elicitation"
],
"timeout": 86400000,
"autoApprove": [
"interactive_feedback"
],
"env": {
"FEEDBACK_TEMPLATE": "=== User Feedback ===\n{{feedback}}", // Optional: Adjust the feedback message template
"FEEDBACK_PROMPT": "Custom tool description for the AI assistant to understand its purpose and rules" // Optional: Customize the tool prompt description
}
}
Environment Variables
FEEDBACK_TEMPLATE
Controls the output format template for the feedback. Use {{feedback}}
as a placeholder for the actual user feedback content.
Example Configuration:
"FEEDBACK_TEMPLATE": "š Feedback Content: {{feedback}}\n\nā
Please handle the related matters promptly."
FEEDBACK_PROMPT
Customizes the tool's description, which the AI assistant uses to understand the tool's purpose and usage rules.
Multi-language Support Example:
// English Environment
"FEEDBACK_PROMPT": "Interactive feedback collection tool. Call this tool to collect user feedback during task execution. Must call repeatedly until user explicitly says 'end'."
// Chinese Environment
"FEEDBACK_PROMPT": "äŗ¤äŗå¼åé¦ę¶éå·„å
·ćåØä»»å”ę§č”čæēØäøč°ēØę¤å·„å
·ę¶éēØę·åé¦ćåæ
é”»éå¤č°ēØē“å°ēØę·ę甮蔨示ē»ęć"
// Japanese Environment
"FEEDBACK_PROMPT": "ć¤ć³ćæć©ćÆćć£ććć£ć¼ććććÆåéćć¼ć«ććæć¹ćÆå®č”äøć«ćć®ćć¼ć«ćå¼ć³åŗćć¦ć¦ć¼ć¶ć¼ćć£ć¼ććććÆćåéćć¾ććć¦ć¼ć¶ć¼ćę確ć«ēµäŗć示ćć¾ć§ē¹°ćčæćå¼ć³åŗćåæ
č¦ćććć¾ćć"
Contributing
Contributions are welcome! Please see the file for details on how to contribute to this project.
License
This project is licensed under the MIT License. See the file for details.
Disclaimer
This project was generated by an AI and has not undergone complete testing and optimization. Please assess the risks before using it. The author assumes no responsibility.