{
  "id": 6093898,
  "title": "From AI Solutions to Shared Knowledge: Building an MCP for the Community",
  "url": "https://urgent.news/2026/09/07/from-ai-solutions-to-shared-knowledge-building-an-mcp-for-the",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-07T03:34:19.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/pascal_cescato_692b7a8a20/from-ai-solutions-to-shared-knowledge-building-an-mcp-for-the-community-6bk"
  },
  "original_language": "en",
  "account": "This submission for the Weekend Challenge focuses on the Shared Knowledge MCP, an innovative project that transforms AI-assisted problem solutions into valuable community knowledge. The key idea is to prevent the loss of useful information that occurs when AI conversations end. Instead of simply providing an answer, Shared Knowledge MCP converts the solution into a Markdown article and submits it as a GitHub Pull Request for human review. Only after the Pull Request is merged does the article become publicly available on a documentation site, accompanied by an audio version generated with ElevenLabs.\n\nThe project prioritizes explicit user consent for sharing knowledge, ensuring that private conversations remain confidential. The MCP server extracts relevant solutions from AI conversations, structures them as standalone Markdown articles, validates them, and submits them as GitHub Pull Requests. The review process is human-only, preventing automatic publication of any content.\n\nThe demo of Shared Knowledge MCP showcases a real-world implementation through two different approaches. First, a script executed the entire process, resulting in a real Pull Request that addressed a technical problem. The PR went through various stages, including validation by the MCP server, an automated review by GitHub Copilot, and a merge into the main branch. Following the merge, an audio version of the article was generated and published on the documentation site.\n\nSecond, a real AI client, GitHub Copilot Chat in Agent mode, was used to test the system. This assistant spontaneously searched the existing knowledge base for duplicates before deciding to publish a new contribution. This behavior was not hard-coded into the MCP server but rather demonstrated the flexibility and adaptability of the system.\n\nThe Shared Knowledge MCP project is open source on GitHub, allowing community members to contribute and build upon the existing knowledge base. The core principle of this project is to encourage users to not only receive AI assistance but also to actively share their solutions, fostering a collaborative environment for solving technical problems efficiently and effectively.",
  "summary": "This is a submission for the Weekend Challenge: Generosity Edition Don't Just Ask AI. Give the Answer Back. AI is a real force multiplier for software development. It's also the ideal companion for solving technical problems fast. But all that knowledge — we keep it to ourselves. Or rather, we lose it. The story always stops there. Question → answer → problem solved — and the conversation sinks…",
  "key_points": [
    "Shared Knowledge MCP transforms AI solutions into Markdown articles",
    "GitHub Pull Requests require human review before public publication",
    "Project prioritizes user consent to protect private conversations"
  ],
  "editors_take": "This development transforms AI-assisted problem solutions into community knowledge, preventing information loss and promoting collaboration by converting solutions into publicly available documentation with human review and validation.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}