Back to skill

Security audit

alibabacloud-sysagent

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because its installation guide reportedly runs a downloaded shell script without verifying it; the privacy warning finding adds concern, and scan coverage was partial.

Review the installation steps before proceeding. Consider whether you trust the source of the downloaded script and can verify it before running it. Avoid sending credentials, secrets, or unnecessary sensitive details in diagnostic requests. The scan was partial: five of six components were fully inspected, and one static analysis was degraded, so the evidence is incomplete.

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
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

The workflow sends the user's natural-language goal to Alibaba Cloud without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

Step 1 instructs the agent to submit the user's goal to SysAgent, which transmits user-provided text to Alibaba Cloud. Although the document identifies the cloud service, it does not warn that submitted diagnostic requests may contain sensitive information or advise excluding secrets and unnecessary private data.

Content

No source excerpt is available for this finding.

The installation workflow executes a downloaded shell script without warning about remote code execution.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The command immediately executes a remotely downloaded script without inspecting it or verifying an independently trusted signature or digest. HTTPS and the official vendor endpoint reduce exposure but do not prevent a compromised distribution endpoint from supplying arbitrary code, which would run with the invoking user's privileges on a potentially sensitive operations host.

Content

No source excerpt is available for this finding.