ClawHub
Back to skill

Security audit

aws-infra

Security checks for vulnerabilities and agentic risk

Overview

This AWS inspection skill is mostly purpose-aligned, but it gives agents access to sensitive AWS credential workflows while under-warning about secrets and broadening beyond read-only inspection.

Review before installing. Use least-privilege read-only AWS roles, prefer AWS SSO or short-lived role credentials, do not paste real access keys into chats or shared terminals, and avoid letting the agent echo, log, or save secrets. Treat AWS command outputs as sensitive even when they are read-only.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Credential Access

High
Category
Privilege Escalation
Confidence
96% confidence
Finding

Referencing ~/.aws/credentials in combination with inline secret examples encourages direct interaction with a well-known credential store. In the context of a skill with exec and write capabilities, this is more dangerous because an agent may be induced to read, modify, or expose credential material from that location.

Content

Scanner excerpt · SKILL.md (reported line 270)May include surrounding context.

md
export AWS_PROFILE=production
aws ec2 describe-instances
# ...
# 配置多个Profile (在~/.aws/credentials中)
# [production]
# aws_access_key_id = AKIAXXXXXXXX
# aws_secret_access_key = YOUR_API_KEY

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The security section claims outputs will not contain sensitive data, yet elsewhere the document includes examples for configuring AWS secrets and temporary credentials. This creates a misleading safety guarantee that may cause operators or agents to surface, log, or persist credentials under the false assumption that the workflow is non-sensitive.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

This markdown file contains user-facing instructions and descriptions primarily in Chinese, while also referencing multiple general-purpose AI agent platforms. Because the skill does not offer an English/localization choice or explain that it is intended only for Chinese-speaking users, it effectively forces a language/locale on users without opt-in.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The documentation shows AWS access keys and secret keys being placed directly into ~/.aws/credentials examples. Even if illustrative, this normalizes storing long-lived secrets in plaintext files and may prompt users or agents to create, copy, or expose credentials in logs, chat transcripts, repositories, or shared environments.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The MFA instructions tell the user/agent to configure returned temporary credentials, including secret keys and session tokens, into a profile. Temporary credentials are still sensitive; documenting this flow without redaction and handling safeguards increases the risk they are copied into logs, terminal history, or persisted files accessible to other processes or users.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is presented as an AWS read-only infrastructure inspection tool, but this section broadens scope into generic file processing, external API integration, and sandboxed command execution. That scope expansion weakens trust boundaries and can cause an agent or user to invoke capabilities unrelated to the declared purpose, increasing the chance of unsafe command execution or data handling.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.