Terracotta AI

Terracotta AI is a specialized code review tool that focuses exclusively on infrastructure as code (IaC) pull requests. It hooks into your GitHub workflow, automatically scanning Terraform, Pulumi, and CloudFormation templates for security misconfigurations, policy violations, and potential drift. Unlike generic AI code reviewers, Terracotta understands cloud resource semantics — it knows that a public S3 bucket is dangerous, that missing encryption on an RDS instance is a compliance risk, and that overly permissive IAM roles break the principle of least privilege. The tool is built for DevOps engineers, platform teams, and security-conscious infrastructure engineers who want to catch issues early without adding manual toil. It works as a GitHub app that posts PR comments with clear explanations and remediation suggestions. You can define custom policies using a YAML-based rules engine, or rely on Terracotta's built-in library of hundreds of cloud provider-specific checks derived from CIS benchmarks and AWS Well-Architected best practices. What makes Terracotta different is its deep domain focus. While general-purpose AI assistants might spot syntax errors, Terracotta understands intent — it can detect when a security group rule is overly broad, when a Terraform module is being deprecated, or when cost-optimization opportunities exist (like right-sizing instance types). It also keeps a memory of past review patterns across your repos, reducing noise and learning what your team considers acceptable. Terracotta operates on a freemium model: individual developers get 100 free reviews per month, while teams pay per seat. The platform supports private repos and offers a self-hosted option for air-gapped environments. It does not store your IaC files beyond the duration of the review, and all scanning happens in ephemeral containers.