Hackernews posts about Simon Willison

  1. GitHub MCP exploited: Accessing private repositories via MCP (invariantlabs.ai)
    508 points by andy99 21 days ago | 297 comments
  2. Show HN: Sllm.nvim – Integrate Simon’s LLM cli into Neovim (500 LOC Lua) (github.com)
    4 points by mozanunal 26 days ago | 1 comments
  3. Show HN: llm.sh – LLMs in the command line in 45 lines of bash (github.com)
    2 points by jakobfrick 10 days ago | discuss
  4. The last six months in LLMs, illustrated by pelicans on bicycles (simonwillison.net)
    950 points by swyx 9 days ago | 232 comments
  5. Gemini Diffusion (simonwillison.net)
    890 points by mdp2021 26 days ago | 244 comments
  6. Claude 4 System Card (simonwillison.net)
    700 points by pvg 23 days ago | 252 comments
  7. Show HN: My LLM CLI tool can run tools now, from Python code or plugins (simonwillison.net)
    529 points by simonw 20 days ago | 165 comments
  8. GitHub issues is almost the best notebook in the world (simonwillison.net)
    314 points by ingve 22 days ago | 195 comments
  9. Highlights from the Claude 4 system prompt (simonwillison.net)
    309 points by Anon84 21 days ago | 90 comments
  10. Design Patterns for Securing LLM Agents Against Prompt Injections (simonwillison.net)
    109 points by simonw 3 days ago | 27 comments
  11. Building software on top of large language models (simonwillison.net)
    36 points by gregorymichael 26 days ago | discuss
  12. Why are front end dev demand so high if front end development is easier? (2012) (simonwillison.net)
    34 points by thunderbong 10 days ago | 31 comments
  13. 105 Vibe-Coded Tools (tools.simonwillison.net)
    25 points by azhenley 8 days ago | 3 comments
  14. I don't like ChatGPT's new memory dossier (simonwillison.net)
    24 points by ingve 26 days ago | 6 comments
  15. I don't like ChatGPT's new memory dossier (simonwillison.net)
    13 points by Sami_Lehtinen 26 days ago | 2 comments
  16. How Often Do LLMs Snitch? Recreating Theo's SnitchBench with LLM (simonwillison.net)
    12 points by Philpax 16 days ago | 4 comments
  17. Notes on – Anthropic: How we built our multi-agent research system (simonwillison.net)
    12 points by thunderbong 1 day ago | discuss
  18. 'How come I can't breathe?': Musk's data company draws a backlash in Memphis (simonwillison.net)
    10 points by gregdoesit 4 days ago | discuss
  19. No build front end is so much more fun (simonwillison.net)
    9 points by xnx 16 days ago | 1 comments
  20. Highlights from the Claude 4 system prompt (simonwillison.net)
    9 points by dcre 22 days ago | 1 comments
  21. Claude-Trace (simonwillison.net)
    9 points by tosh 14 days ago | discuss
  22. Remote Prompt Injection in Gitlab Duo Leads to Source Code Theft (simonwillison.net)
    9 points by thunderbong 24 days ago | discuss
  23. The last six months in LLMs, illustrated by pelicans on bicycles (simonwillison.net)
    7 points by simonw 10 days ago | discuss
  24. The Perfect Commit (2022) (simonwillison.net)
    7 points by saikatsg 15 days ago | discuss
  25. Quoting Neal Stephenson (simonwillison.net)
    6 points by mdp2021 29 days ago | 3 comments
  26. An agent is an LLM wrecking its environment in a loop (simonwillison.net)
    6 points by gk1 11 days ago | discuss
  27. Honey Badger (simonwillison.net)
    6 points by tosh 24 days ago | discuss
  28. AI agent lethal trifecta: private data, untrusted content, exfiltration vectors (simonwillison.net)
    5 points by simonw about 7 hours ago | 1 comments
  29. 'How come I can't breathe?': Musk's data company draws a backlash in Memphis (simonwillison.net)
    4 points by diggan 3 days ago | 1 comments
  30. OpenAI slams court order to save all ChatGPT logs, including deleted chats (simonwillison.net)
    3 points by huerlisi1 11 days ago | 1 comments
More