mcpField
C
cloudinfra

cloudinfra

Local
github.com/CloudCorpRecordsCloud & Infra0Updated 2w ago
Repository

About cloudinfra

cloudinfra is an MCP server for AWS infrastructure practitioners, offering Terraform Registry documentation and module lookups, plan policy scans, and explanations of resource types. Its policy scanner runs locally and deterministically; it does not run Terraform operations or deploy infrastructure.

Key features

  1. 01Search Terraform provider docs and versions
  2. 02Find modules and view details
  3. 03Scan Terraform plan policies
  4. 04Explain common AWS resource types

Use cases

  1. 01Find provider documentation and modules
  2. 02Check plans for public S3 and open security groups
  3. 03Detect unencrypted storage and wildcard IAM rules

MCP support

It exposes 10 tools as a stdio MCP server, letting MCP clients query the Terraform Registry, scan plan files, and get AWS resource explanations.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "cloudcorprecords-cloudinfra": {
      "command": "npx",
      "args": [
        "-y",
        "@reneturcios/cloudinfra"
      ]
    }
  }
}
Run commandnpx -y @reneturcios/cloudinfra

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@reneturcios/cloudinfra0.1.0
ocighcr.io/cloudcorprecords/cloudinfra:0.1.0

Tools

10

Saved tool list · updated 1w ago

ParameterTypeRequiredDescription
querystringYesSearch text to match against Terraform provider documentation topics.
providerstringNoOptional provider name to restrict the documentation search to one provider.

Details

From the project README

cloudinfra

AWS infrastructure-as-code guidance as MCP tools: Terraform Registry documentation and module lookups, a local policy scan for terraform plan files, and plain-language explanations of common AWS resource types. cloudinfra is a clean-room rebuild of the core surface of the aws-terraform MCP server, under a new name: fresh code and prose, no copied branding.

Tools (core subset, 10)

search_provider_docs, get_provider_doc_details, search_cc_provider_docs, list_provider_resources, search_modules, get_module_details, get_latest_provider_version, search_user_guide, run_policy_scan, explain_resource.

Not in this subset: Terraform/Tofu CLI execution (init/plan/apply), provider schema generation, pricing and cost estimation, LLM-assisted code or module generation, state manipulation, and any write or deploy operation.

How it works

The docs and module tools are shaped GET requests against the public Terraform Registry API (https://registry.terraform.io/v1, base overridable with TERRAFORM_REGISTRY_URL) through an injectable fetch function. No credentials are needed for any tool.

run_policy_scan is fully local and deterministic. Give it a terraform plan JSON (the output of terraform show -json) and it walks planned_values (including child modules; resource_changes is a fallback) checking four rule families:

  • public-s3 — S3 buckets with a public ACL
  • unencrypted-storage — RDS/EBS without encryption, S3 buckets with no server-side encryption configuration
  • wildcard-iam — IAM policies granting wildcard actions or resources
  • open-security-group — ingress open to 0.0.0.0/0 or ::/0

An external scanner can be injected through the library API as a scan executor; none is bundled — the local checker is the default and the only implemented scanner.

explain_resource answers from a small curated catalogue of common AWS resource types written for this rebuild, and says so plainly for types outside the catalogue.

CLI

npx @reneturcios/cloudinfra explain_resource '{"resource_type":"aws_s3_bucket"}'
npx @reneturcios/cloudinfra run_policy_scan '{"plan":{...}}'
npx @reneturcios/cloudinfra mcp        # stdio MCP server

Each tool name is a subcommand taking one JSON object argument.

MCP client config

{
  "mcpServers": {
    "cloudinfra": {
      "command": "npx",
      "args": ["-y", "@reneturcios/cloudinfra", "mcp"]
    }
  }
}

Docker

docker run -i ghcr.io/cloudcorprecords/cloudinfra

Development

npm install
npm run build
npm test

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.