HN Explorer

dheerajmp

↗ HN profile [ 72.7 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Claude Skills for Cybersecurity (github.com)
    3 points 0
  2. Gstack to Be Renamed as Gslop (twitter.com)
    3 points 0
  3. Claude Code Leaks (github.com)
    4 points 1
  4. Claude Code full source code leaked on NPM (github.com)
    47 points 6
  5. Non trivial client-side UI manipulation in runtime using AI (leanmcp.com)
    1 points 0
  6. Trivy Attacks (ramimac.me)
    3 points 0
  7. Delve.co and SoC Type 2 Compliance Scandal and Audit (kaggle.com)
    2 points 0
  8. Laws of UX – in the age of AI vibe coding (lawsofux.com)
    2 points 0
  9. Never knew news websites could survive without ads (text.npr.org)
    9 points 8
  10. Awesome-Webmcp (github.com)
    3 points 3
  11. Infinite Mario levels – generated on the fly (supermario.leanmcp.live)
    2 points 2
  12. Building AI agent for our own company (blog.leanmcp.com)
    2 points 1
  13. WebMCP App for Composing Music (music.leanmcp.live)
    10 points 5
  14. Mathematicians issue a major challenge to AI (scientificamerican.com)
    2 points 1
  15. I went through the OpenClaw Source code. And here are my observations (pai.dev)
    1 points 1
  16. 1 points 0
  17. Hate filling forms – Built an AI that just does that in one click (chromewebstore.google.com)
    2 points 1
  18. Open sourcing my SSE virtualization layer for Remote MCP (github.com)
    3 points 1
  19. Show HN: Built a No-code platform to build and deploy MCP Servers (ship.leanmcp.com)

    Most of the LLMs behave like crap when vibecoding for MCPs. Either the libraries used are totally outdated or confuses MCP with "Microsoft Certified Professional". I built this during the MCP hackathon by YC. And got…

    1 points 0
  20. 1 points 0
  21. Show HN: LLM Labs – UI to Test 3 LLMs in Parallel (labs.fprime.ai)

    Try out multiple LLMs in parallel. LLama 3, DBRX, Grok (X.AI), Sonar, llama.cpp. Run parallely on AWS, Together AI and Groq.

    1 points 33