AGX: The Verifiable AI Workflow Engine for DevOps

Verification First Terraform

AGX uses static analysis to validate LLM-generated plans before execution. It compiles natural language into verified Terraform, ensuring type safety and logical correctness.

This web app shows an early prototype version of the engine. An updated version with advanced features will power an open-source CLI (Q1 2026).

GitHub

Describe your infrastructure...

Python to TF Script

About the Engine

Live in this Demo:

Uses a deterministic validation stage to enforce strict type safety. The engine checks every step against a function registry to prevent hallucinated parameters or invalid dependencies.

Open Source CLI (Q1 2026):

Evolving into a graph-based execution engine powered by a custom lightweight model.

Available tools (demo subset):

  • set_bucket_name
  • create_aws_s3_bucket
  • aws_s3_bucket_public_access_block
  • save_hcl_to_file
  • sanitise_resource_name
  • combine_two_hcl_blocks

AGX CLI (private alpha)

  • Cloud Engine: Powered by a custom 12B model.
  • Local Execution: Your AWS credentials never leave your machine.
  • Open Source: Fully auditable CLI code (coming Q1).
Built by Ammar Qureshi, Founder of AGX