AWS‑IReveal‑MCP
# AWS‑IReveal‑MCP
**AWS‑IReveal‑MCP** is a Model Context Protocol (MCP) server designed to give security teams and incident responders a unified interface to AWS services useful for investigation. By connecting AWS‑IReveal‑MCP to any MCP client (such as Claude Desktop or Cline), you can invoke queries and analyses across multiple AWS services without leaving your LLM‑driven workspace.
## Features
AWS‑IReveal‑MCP integrates with the following AWS services and functionalities:
- **CloudTrail** — Management event logs for API activity
- **Amazon Athena** — SQL queries over CloudTrail logs
- **CloudWatch** — Operational logs and ad hoc analysis
- **Amazon GuardDuty** — Threat detection and finding investigation
- **AWS Config** — Resource configuration history and compliance status
- **VPC Flow Logs** — Network traffic metadata for forensic analysis
- **Network Access Analyzer** — Reachability checks across SG/NACL/VPC
- **IAM Access Analyzer** — Policy and resource‑based access findings
Together, these services let you
- Trace “who did what, when, and where” (CloudTrail, Config)
- Examine detailed data events (Athena)
- Search and visualize logs (CloudWatch, VPC Flow Logs)
- Surface security alerts (GuardDuty, IAM Access Analyzer)
- Verify network reachability and configuration (Network Access Analyzer)
### Example Prompts
- analyze activity by IP x.x.x.x in the last 5 days
- analyze activity by role 'sysadmin' in the last 24 hours
- investigate suspicious activity on cloudtrail in the last 7 days on us-west-2
- is there any data event on buckets with name containing 'customers' in the last 7 days?
- investigate cloudwatch logs related to Bedrock
- propose remediations for GuardDuty findings with high risk happened in the last 2 days
- identify non-compliant resources, explain violated rules, and suggest remediation
## Installation
### Prerequisites
- Install UV with:
```bash
curl -Ls https://astral.sh/uv/install.sh | sh
```
- Clone the repo and set up the environment (this will install the required dependencies):
```bash
git clone https://github.com/Brucedh/aws-ireveal-mcp.git
cd aws-ireveal-mcp
uv venv
source .venv/bin/activate
```
### Configuration
Add the following configuration to your MCP client's settings file:
```
{
"mcpServers": {
"aws-ireveal": {
"command": "uv",
"args": [
"--directory",
"/path_to_your/aws-ireveal-mcp",
"run",
"server.py"
],
"env": {
"AWS_PROFILE": "<YOUR_PROFILE>"
}
}
}
}
```TDQS
Scored across 25 tools
Most tools are clearly distinct by AWS service and operation, but there is some overlap between cloudtrail_lookup_events and athena_query_events, as both query CloudTrail logs with different capabilities. The descriptions help clarify when to use each, but an agent might initially be confused about which tool to select for certain log queries.
All tool names follow a consistent pattern of service_verb_noun (e.g., accessanalyzer_get_analyzer, cloudtrail_describe_trails). This makes it easy to understand the service and action at a glance, with no deviations in naming conventions across the set.
With 25 tools, the count feels heavy for a single server, but it covers multiple AWS security and monitoring services (Access Analyzer, CloudTrail, Athena, CloudWatch, Config, EC2, GuardDuty, Network Insights). While comprehensive, it may overwhelm agents with its breadth, bordering on too many for a cohesive set.
The tool set provides good coverage for security analysis across AWS services, with CRUD-like operations for each service (e.g., list, get, describe, query). Minor gaps exist, such as missing update or delete operations for some resources, but agents can perform core workflows without major dead ends.