Generate AWS IAM policies via chat interface with GPT-4 assistant
Ready for 3.0Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.
Nodes used (7)
- Chat Trigger
@n8n/n8n-nodes-langchain.chatTrigger - Simple Memory
@n8n/n8n-nodes-langchain.memoryBufferWindow - OpenAI Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenAi - Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured - AI Agent
@n8n/n8n-nodes-langchain.agent - HTTP Request
n8n-nodes-base.httpRequest - Send Email
n8n-nodes-base.emailSend
What it does
Chat-Based AWS IAM Policy Generator with OpenAI Agent > Chat-driven workflow that lets IT and DevOps teams generate custom AWS IAM policies via AI, automatically apply them to AWS, and send an email notification with policy details. 👤 Who’s it for This workflow is designed for: - Cloud Engineers / DevOps who need to quickly generate and apply custom IAM policies in AWS. - IT Support / Security teams who want to create IAM policies through a chat-based interface without manually writing JSON. - Teams that want automatic notifications (via email) once new policies are created. ⚙️ How it works / What it does 1. Trigger → Workflow starts when a chat message is received. 2. IAM Policy Creator Agent → Uses OpenAI to: - Interpret user requirements (e.g., service, actions, region). - Generate a valid IAM policy JSON following AWS best practices. 3. IAM Policy HTTP Request → Sends the generated policy to AWS IAM CreatePolicy API. 4. Email Notification → Once AWS responds with a CreatePolicyResponse, an email is sent with policy details (name, ARN, ID, timestamps, etc.) using n8n mapping. Result: The user can chat with the AI agent, create a policy, and receive an email confirmation with full details. 🛠 How to set up 1. Chat Trigger Node - Configure the When chat message received node to connect your preferred chat channel (Slack, MS Teams, Telegram, etc.). 2. IAM Policy Creator Agent - Add OpenAI Chat Model as the LLM. - Use a system prompt that enforces AWS IAM JSON best practices (least privilege, correct JSON structure). - Connect Memory (Simple Memory) and Structured Output Parser to ensure consistent JSON output. 3. IAM Policy HTTP Request - Set method: POST - URL: https://iam.amazonaws.com/ - Add authentication using AWS Signature v4 (Access Key + Secret Key). - Body: - Action=CreatePolicy - PolicyName={{ $json.CreatePolicyResponse.CreatePolicyResult.Policy.PolicyName }} - PolicyDocument={{ $json.policyDocument }} - Version=2010-05-08 4. Email for tracking 📋 Requirements - n8n instance (self-hosted or cloud). - AWS IAM user/role with permission to iam:CreatePolicy. - AWS Access Key + Secret Key (for SigV4 signing in HTTP request). - OpenAI API key (for the Chat Model). - Email server credentials (SMTP or provider integration). 🎨 How to customize the workflow - Restrict services/actions → Adjust the IAM Policy Creator Agent system prompt to limit what services/policies can be generated. - Notification channels → Replace the email node with Slack, MS Teams, or PagerDuty to alert other teams. - Tagging policies → Modify the HTTP request to include Tags when creating policies in AWS. - Human-readable timestamps → Add a Function or Set node to convert CreateDate and UpdateDate from Unix epoch to ISO datetime before sending emails. - Approval step → Insert a manual approval node before sending the policy to AWS for compliance workflows.