mcp-tfstate-reader
A local MCP server that parses Terraform .tfstate files and lets AI agents audit enterprise infrastructure for security misconfigurations without requiring direct cloud credentials.
README
mcp-tfstate-reader
A local Model Context Protocol (MCP) server that parses Terraform .tfstate files and lets AI agents audit enterprise infrastructure for security misconfigurations — without requiring direct cloud credentials.
Why?
Terraform state files are the single source of truth for what's actually deployed in your cloud. Security teams need to audit them regularly, but the traditional workflow is manual: run CLI tools, parse terminal output, repeat.
mcp-tfstate-reader gives AI agents (like Claude) structured, read-only access to your Terraform state. Instead of copying JSON into a chat window, you ask a question and the agent calls the right tool automatically.
Features
| Tool | Description |
|---|---|
list_resources |
Parse a .tfstate file and list every managed resource with its address and type |
audit_security |
Scan for common misconfigurations (see below) |
get_resource_detail |
Dump the full attributes of a specific resource by address |
summarize_state |
High-level overview: resource counts by type/module, providers, tags, regions |
compare_states |
Infrastructure drift detection — diff two state files for added/removed/modified |
Security checks in audit_security
- S3 — buckets without server-side encryption; buckets without versioning enabled
- S3 ACL — bucket ACLs set to
public-readorpublic-read-write - Security Groups — ingress rules open to
0.0.0.0/0on sensitive ports: 22 (SSH), 3389 (RDP), 5432 (PostgreSQL) - IAM — policies with wildcard
*actions (full admin access) - RDS — instances without
storage_encrypted = true; publicly accessible instances - EBS — volumes without encryption
- EC2 — instances with
associate_public_ip_address = true - Lambda — functions not deployed in a VPC
- KMS — keys without automatic key rotation enabled
- ElastiCache — replication groups without transit encryption
- SNS — topics without KMS encryption
- SQS — queues without KMS encryption
- ALB/NLB — load balancers without access logs enabled
- CloudWatch — log groups without a retention policy
Quick Start
1. Install
pip install mcp-tfstate-reader
2. Configure Claude Desktop
Add this to your Claude Desktop MCP configuration file:
| OS | Path |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows | %APPDATA%\Claude\claude_desktop_config.json |
Recommended — with uvx (no install required):
{
"mcpServers": {
"tfstate-reader": {
"command": "uvx",
"args": ["mcp-tfstate-reader"]
}
}
}
Note: Claude Desktop may not inherit your terminal's
$PATH. If the server fails to connect, use the absolute path touvx(find it withwhich uvxin your terminal):{ "mcpServers": { "tfstate-reader": { "command": "/full/path/to/uvx", "args": ["mcp-tfstate-reader"] } } }
Alternative — installed via pip:
{
"mcpServers": {
"tfstate-reader": {
"command": "mcp-tfstate-reader"
}
}
}
Alternative — from source (virtualenv):
{
"mcpServers": {
"tfstate-reader": {
"command": "/absolute/path/to/mcp-tfstate-reader/.venv/bin/python",
"args": ["-m", "mcp_tfstate_reader.server"]
}
}
}
3. Restart Claude Desktop
Fully quit (Cmd+Q on macOS) and reopen. Look for the tools icon to confirm the server is connected.
4. Ask a question
"Audit the Terraform state file at
/path/to/terraform.tfstatefor security issues."
Example interaction
Claude autonomously chains the tools — listing resources first, running the audit, then drilling into critical findings:
Found 17 finding(s):
[CRITICAL] aws_iam_policy.admin: IAM policy contains a wildcard (*) action — grants unrestricted permissions.
[HIGH] aws_s3_bucket.assets: S3 bucket has no server-side encryption configuration.
[HIGH] aws_s3_bucket_acl.assets: S3 bucket ACL is set to 'public-read' — allows public access.
[HIGH] aws_security_group.bastion: Security group allows 0.0.0.0/0 ingress on port 22.
[HIGH] aws_security_group.rdp_open: Security group allows 0.0.0.0/0 ingress on port 3389.
[HIGH] aws_db_instance.prod: RDS instance storage is not encrypted.
[HIGH] aws_db_instance.prod: RDS instance is publicly accessible.
[HIGH] aws_ebs_volume.data: EBS volume is not encrypted.
[HIGH] aws_elasticache_replication_group.sessions: ElastiCache replication group does not have transit encryption enabled.
[MEDIUM] aws_s3_bucket.assets: S3 bucket does not have versioning enabled.
[MEDIUM] aws_instance.web: EC2 instance has a public IP address assigned.
[MEDIUM] aws_lambda_function.processor: Lambda function is not deployed in a VPC.
[MEDIUM] aws_kms_key.app: KMS key does not have automatic key rotation enabled.
[MEDIUM] aws_sns_topic.alerts: SNS topic is not encrypted with a KMS key.
[MEDIUM] aws_sqs_queue.jobs: SQS queue is not encrypted with a KMS key.
[MEDIUM] aws_lb.frontend: Load balancer does not have access logs enabled.
[MEDIUM] aws_cloudwatch_log_group.app: CloudWatch log group has no retention policy (logs kept indefinitely).
Troubleshooting
Server not appearing in Claude Desktop
- Ensure Claude Desktop is fully restarted (quit with
Cmd+Q, not just close the window). - Check your config JSON is valid — a trailing comma or typo will silently break it.
- Use absolute paths if
uvxormcp-tfstate-readeraren't found.
"File not found" errors
The tool reads files from your local filesystem. Use the full absolute path (e.g. /Users/you/infra/terraform.tfstate), not relative paths.
This is Claude Desktop only
MCP servers work with the Claude Desktop app, not claude.ai in your browser. The web interface does not have access to local MCP servers or your filesystem.
Development
# Clone and set up
git clone https://github.com/berkayildi/mcp-tfstate-reader.git
cd mcp-tfstate-reader
make setup
# Run tests
make test
# Build distribution
make build
# Run the server locally (stdio)
make start
# Clean everything
make clean
License
MIT © Berkay Yildirim
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.