<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: GAURANG UDGIRKAR</title>
    <description>The latest articles on DEV Community by GAURANG UDGIRKAR (@gaurang_udgirkar_1033a7d8).</description>
    <link>https://dev.to/gaurang_udgirkar_1033a7d8</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1946040%2F60acf140-4b54-437e-ae7f-e4394aff0593.png</url>
      <title>DEV Community: GAURANG UDGIRKAR</title>
      <link>https://dev.to/gaurang_udgirkar_1033a7d8</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/gaurang_udgirkar_1033a7d8"/>
    <language>en</language>
    <item>
      <title>GhostPR : MCP For Future</title>
      <dc:creator>GAURANG UDGIRKAR</dc:creator>
      <pubDate>Sun, 14 Jun 2026 10:14:18 +0000</pubDate>
      <link>https://dev.to/gaurang_udgirkar_1033a7d8/ghostpr-mcp-for-future-22d4</link>
      <guid>https://dev.to/gaurang_udgirkar_1033a7d8/ghostpr-mcp-for-future-22d4</guid>
      <description>&lt;p&gt;AI coding assistants are becoming very good at reading code.&lt;br&gt;
But they still struggle with something humans rely on constantly:&lt;br&gt;
understanding why the code exists.&lt;br&gt;
Engineering decisions rarely live in the code itself.&lt;br&gt;
They live in pull requests, review comments, issues, RFCs, and team discussions.&lt;br&gt;
Over time, that context disappears.&lt;br&gt;
Months later, an AI revisits the codebase, sees an unusual implementation, labels it as technical debt, and confidently removes it—without knowing the reason it was introduced.&lt;br&gt;
The consequences can be expensive:&lt;br&gt;
❌ Reintroduced bugs&lt;br&gt;
❌ Broken workarounds&lt;br&gt;
❌ Repeated engineering mistakes&lt;br&gt;
❌ Lost organizational knowledge&lt;br&gt;
To explore this problem, we built GhostPR.&lt;br&gt;
&lt;em&gt;GhostPR&lt;/em&gt; is a decision-memory layer for agentic IDEs.&lt;br&gt;
It captures engineering decisions, links them to code, and retrieves historical context before AI-generated changes are applied.&lt;br&gt;
To make memory useful—not permanent—each decision stores:&lt;br&gt;
• Decision&lt;br&gt;
• Reasoning&lt;br&gt;
• Evidence source (PR / Issue / Review / RFC)&lt;br&gt;
• Confidence score&lt;br&gt;
• Current status (Active / Questionable / Deprecated)&lt;br&gt;
Example:&lt;br&gt;
Developer:&lt;br&gt;
Refactor auth/session.ts&lt;br&gt;
GhostPR:&lt;br&gt;
⚠ Historical Context&lt;br&gt;
Decision: Custom OAuth refresh flow&lt;br&gt;
Reason: Provider incompatibility with standard refresh&lt;br&gt;
Evidence: PR #184 discussion&lt;br&gt;
Outcome: Authentication failures reduced after deployment&lt;br&gt;
Status: Active&lt;br&gt;
Recommendation: Review before modifying token expiry logic.&lt;br&gt;
The goal isn’t to tell developers what to do.&lt;br&gt;
It’s to make sure both humans and AI can access the reasoning behind critical engineering decisions.&lt;br&gt;
Because understanding why code exists is often more valuable than understanding what it does.&lt;br&gt;
We’re continuing to improve this and I’d genuinely love feedback, critiques, ideas, or contributions from people thinking about agent memory, developer tools, and AI workflows.&lt;br&gt;
If this resonates with you, let’s build better engineering memory together.&lt;br&gt;
hashtag#AI hashtag#AgenticAI hashtag#DeveloperTools hashtag#SoftwareEngineering hashtag#MCP hashtag#AgentMemory hashtag#OpenSource hashtag#GhostPR&lt;/p&gt;

&lt;p&gt;Github Repo: &lt;a href="https://github.com/Gaurang774/GhostPR" rel="noopener noreferrer"&gt;https://github.com/Gaurang774/GhostPR&lt;/a&gt;&lt;/p&gt;

</description>
      <category>agents</category>
      <category>ai</category>
      <category>mcp</category>
      <category>showdev</category>
    </item>
  </channel>
</rss>
