Aws Networking Audit
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This appears to be a coherent read-only AWS networking audit skill, but it needs scoped AWS credentials and has under-declared CLI/MCP requirements users should verify.
This skill looks appropriate for a read-only AWS VPC networking audit. Before installing or using it, confirm you are comfortable granting AWS read access, use a narrowly scoped read-only role, specify the target account/region/VPC, keep Flow Log queries time-bounded, and verify whether the declared aws-network-mcp dependency is approved in your environment.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
If used with broad AWS credentials, the agent can read detailed cloud network configuration and traffic-log metadata across the granted scope.
The skill requires AWS account credentials and read-only permissions that can inspect VPC configuration, CloudWatch Flow Logs, and potentially multiple accounts; this is expected for the stated audit purpose but sensitive.
AWS CLI v2 configured with valid credentials (`aws sts get-caller-identity` succeeds) ... IAM permissions ... `logs:FilterLogEvents`, `logs:DescribeLogGroups` ... Multi-account audits require cross-account IAM roles or AWS Organizations access
Use a least-privilege read-only role, limit the account/region/VPC scope, and avoid granting admin or organization-wide access unless explicitly needed.
The agent may run read-only AWS CLI commands that reveal cloud inventory, security groups, route tables, Transit Gateway routes, and Flow Log events.
The skill's main capability is to guide AWS CLI use. The commands are described as read-only and fit the audit purpose, but they still provide broad visibility into AWS networking resources and logs.
All commands are non-modifying (`describe`, `get`, `list`, `search`, `filter`). No command creates, modifies, or deletes resources.
Review the commands before running them, keep queries scoped by region/VPC/log time range, and do not substitute modifying AWS commands into the workflow.
Install-time metadata may not fully warn users that the skill expects AWS tooling, credentials, network egress, and possibly an MCP connector.
The registry-level metadata under-declares requirements that SKILL.md itself declares, including AWS CLI, AWS_ACCESS_KEY_ID, Amazon egress, and an MCP dependency. This appears to be a metadata completeness issue rather than hidden behavior.
Required binaries (all must exist): none ... Required env vars: none ... Primary credential: none ... Capability signals: No capability tags were derived.
Treat the SKILL.md prerequisites as authoritative and verify AWS CLI, credential, egress, and MCP requirements before enabling the skill.
If the MCP dependency is enabled, AWS networking data or tool access could be mediated by that connector.
The skill metadata declares an MCP dependency and AWS egress, but the provided artifacts do not describe that MCP server's identity, permissions, or data boundaries. No code or automatic MCP use is shown.
"mcpDependencies":["aws-network-mcp"],"egressEndpoints":["*.amazonaws.com:443"]
Only enable the MCP dependency if its source and permissions are trusted, and confirm it receives only the data needed for the audit.