Watadot Aws Ec2

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: watadot-aws-ec2 Version: 1.0.0 The watadot-aws-ec2 skill bundle contains standard AWS CLI commands for managing EC2 instances, such as listing, starting, stopping, and configuring security groups. The SKILL.md file provides helpful documentation and follows security best practices, such as recommending IAM roles over hard-coded credentials, with no evidence of malicious intent or prompt injection.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If used with real AWS permissions, these commands can start, stop, or permanently terminate EC2 instances, affecting uptime and costs.

Why it was flagged

The skill documents AWS CLI lifecycle operations, including instance termination. This is aligned with EC2 management and is marked as dangerous, but it can still disrupt workloads if run unintentionally.

Skill content
aws ec2 start-instances --instance-ids <id1> <id2>
aws ec2 stop-instances --instance-ids <id>

# Terminate instance (DANGER)
aws ec2 terminate-instances --instance-ids <id>
Recommendation

Use explicit user approval for lifecycle changes, verify the AWS account and region first, and prefer least-privilege IAM permissions.

What this means

The agent may act with the same AWS privileges as the configured CLI profile, including changing network access rules if that profile allows it.

Why it was flagged

The skill relies on the local AWS CLI and includes account-mutating EC2 and security-group operations. Those commands will use whatever AWS credentials and permissions are configured locally.

Skill content
requires:
  anyBins: [aws]
...
aws ec2 authorize-security-group-ingress --group-id <sg-id> --protocol tcp --port 22 --cidr <your-ip>/32
Recommendation

Run this only with a dedicated AWS profile or role scoped to the needed EC2 actions, and confirm the target account, region, instance IDs, and CIDR ranges before execution.