TesseractSafety
3 inlägg på Tesseracts blogg.
- How an AI agent decides a command is too risky to run How Tesseract checks an agent's shell commands and file writes on 14 frameworks: 13 rules, a second opinion that can only raise, and what happens next. · 15 min läsning
- Who is connected to your computer? Ask, then end it Since 0.11.37 you can ask Tesseract who is connected to your computer, end the session and stop sharing it, all from the chat. Letting anyone in stays yours. · 2 min läsning
- Can the agent delete the wrong folder? Not in Tesseract In Tesseract the agent deletes a file only by its full path, and only one that exists. A blind rm -f * is refused in every mode, Full access included. · 1 min läsning