Mihir Ajmera

CASE STUDIES

Research with the evidence attached

Each study asks one question, answers it with real tools against real systems, and publishes the evidence so you can check the work, including what my own tools got wrong.

  1. OCT 2026 What policy-as-code found in my own Terraform I put three of my own AWS Terraform projects through a compliance gate. Two security Lambdas had permissions that would have broken the fix they exist to apply. 9120
  2. OCT 2026 Red-teaming an LLM, then auditing the red-team 1,280 garak attacks on Llama 3.2 3B, a hardened system prompt, and a release gate. Two of the five headline numbers were wrong until a human re-performed the evidence. 1,28044% → 22%2/5
  3. OCT 2026 What the official MCP READMEs teach you to copy A user access review of the configuration examples in 9 widely used MCP server READMEs. No single one is dangerous; pasted together, they build an exfiltration path. 8/97/91