Back to skill

Security audit

huawei-cloud-dcs-list

Security checks for vulnerabilities and agentic risk

Overview

This read-only cloud inventory skill needs review because exact-name searches silently ignore tag restrictions, and its installer source could not be verified.

Avoid combining exact-name and tag filters until the filtering bug is corrected. Use credentials limited to DCS inventory access, configure secrets outside chat, and protect exports containing infrastructure details. Install KooCLI through Huawei's verified instructions instead of the unverified bundled URL, and select an export path that may safely be overwritten. The bundled code was inspected and the filtering flaw reproduced with mocked data; live cloud behavior and external installer contents were not tested.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (8)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Local exact-name matching is explicitly documented and is not inherently unsafe, but the implementation also discards the requested tag filter without reapplying it locally. This can include instances outside the requested inventory subset in displayed or exported results, although IAM authorization remains enforced.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The supplied command targets a local skill path for scanning; it does not query DCS or process instance data. Although its file location suggests a verification reference, the command's actual function is not among the declared capabilities. This assessment is limited to the supplied chunk and does not establish what other skill files implement.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_read, file_write, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The Markdown output contract mandates Chinese-language headers and status text without user language opt-in.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The verification contract requires exact equality with Chinese column headers and a Chinese file-written message. This enforces a specific output language without offering a locale choice or documenting a justification for the restriction.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/list_dcs.py (reported line 297)May include surrounding context.

python
cmd.extend(extra)
    cmd.extend(argv)
    try:
        proc = subprocess.run(
            cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE,
            timeout=120, universal_newlines=True,
        )

Tainted flow: 'cmd' from os.environ.get (line 291, credential/environment) → subprocess.run (code execution)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/list_dcs.py (reported line 297)May include surrounding context.

python
cmd.extend(extra)
    cmd.extend(argv)
    try:
        proc = subprocess.run(
            cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE,
            timeout=120, universal_newlines=True,
        )

Exact-name mode silently disables the requested server-side name and tag filters.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

When --name-equal is enabled, main passes tags=None and subsequently filters only by name. Instances that do not satisfy the user's tag restriction can consequently appear in output or exported inventory, creating a data-scoping and reporting-integrity issue without bypassing cloud IAM.

Content

No source excerpt is available for this finding.

Acceptance criteria require Chinese-language status and error messages without a user language choice.

Low
Category
Output Handling
Confidence
88% confidence
Finding

The criteria prescribe Chinese text for the empty-region message and the permission-denied hint. These requirements fix the language of user-facing output without documenting user opt-in or a justified language restriction.

Content

No source excerpt is available for this finding.