Skip to content
AIzaadi
اردوStart free
Track

AI Safely

محفوظ اے آئی

8 lessons · 5 hours total

Start track
  1. 1

    Why arbitrary AI execution needs a sandbox

    33 min

  2. 2

    Anthropic Skills — bottled, scoped, repeatable

    34 min

  3. 3

    Code execution sandboxes — E2B, Daytona, Replit, Anthropic

    35 min

  4. 4

    Secrets and keys — what never goes in a prompt

    32 min

  5. 5

    Rate limits, throttling, and graceful queueing

    33 min

  6. 6

    PII handling — redaction, PDPA-2023, and citizen dignity

    36 min

  7. 7

    Prompt injection — input validation and output sanitisation

    38 min

  8. 8

    Capstone — threat model and mitigation plan for one use case

    40 min