Înapoi la articole →

Cel mai recent articol

Ultima verificare

Urmăritori 0 urmăritori
Articole: 23
  1. Disclaimer: Prompt injection is an unsolved problem.

  2. Disclaimer: Prompt injection is an unsolved problem.

  3. Disclaimer: Prompt injection is an unsolved problem.

  4. Disclaimer: Prompt injection is an unsolved problem.

  5. Disclaimer: Prompt injection is an unsolved problem.

  6. Disclaimer: Prompt injection is an unsolved problem.

  7. Disclaimer: This post is for educational purpose, prompt injection is an unsolved problem.

  8. Disclaimer: This post is for educational purpose, prompt injection is an unsolved problem (but running Claude Code in a sandbox would mostly solve this problem).

  9. Disclaimer: This post is for educational purpose, prompt injection is an unsolved problem.

  10. I really wanted to play around with Fable-5...

  11. Another day, another model.

  12. The last (for how long?)

  13. Lastly the openai model gpt-5.4 gained in popularity within the OpenClaw community (maybe this is the reason).

  14. Another BrokenClaw!

  15. Third part of the BrokenClaw saga!

  16. In Dify, I found a cross site scripting vulnerability (XSS) in the web application chat frontend when using echarts.

  17. You remember this experiment with sonnet 4.5?

  18. In the first part of this saga (read first), I described a 0-Click RCE in OpenClaw using prompt injection and exploiting the insecure plugin handling.

  19. This post is not related to OT but I could not resist.

  20. Today we will use dify, an "open-source platform for developing LLM applications" to show how data exfiltration works - a classical problem in Retrieval Augmented Generation (RAG)...

  21. Classic but still up-to-date.

  22. In my last post, I experimented with a python agent and prompt injection using different LLMs.

  23. Inspired by this blog post and out of curiousity, I vibe-built a simple LLM agent in the terminal with a few tools and MCP capability.