Back to skill

Security audit

huawei-cloud-cbr-list

Security checks for vulnerabilities and agentic risk

Overview

The backup-listing code is narrowly scoped, but its installation instructions execute an unverified remote script and its credential-handling assurances conflict.

Use a verified KooCLI download from Huawei's official documentation instead of executing the supplied installer command. Use credentials restricted to backup listing, and understand that SDK fallback reads those keys into process memory despite the main documentation's assurance. The flagged test-file gap was independently investigated; external installer contents and live cloud behavior remain unverified. No malicious payload was found in the bundled files.

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 (7)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, 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 quick start instructs users to download and execute a remote shell script without warning about its system access or potential changes.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The guide downloads a mutable remote installer and immediately executes it without integrity verification or a review step. Although HTTPS protects transport, compromise of the distribution endpoint or installer supply chain could execute arbitrary code with the invoking user's privileges, potentially accessing cloud credentials; the skill's read-only API restriction does not protect this installation path.

Content

No source excerpt is available for this finding.

Pass criteria require Chinese-language error hints without user opt-in.

Medium
Category
Output Handling
Confidence
98% confidence
Finding

The pass criteria explicitly require a Chinese hint on stderr for error paths. The file provides neither a user language choice nor a documented justification for imposing Chinese-language output.

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_cbr_backups.py (reported line 340)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 334, 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_cbr_backups.py (reported line 340)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,
        )

The SDK fallback directly reads AK/SK despite the module's explicit claim that the script never reads credentials itself.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

The module documentation at L007-L009 explicitly states that this script never reads AK/SK itself. However, L435-L452 retrieve access and secret keys directly from environment variables, and L517-L530 pass those values into BasicCredentials. Authentication is justified for the declared SDK fallback, but the assurance that credential handling is delegated entirely to hcloud or the SDK is false.

Content

No source excerpt is available for this finding.

The empty-list acceptance criterion prescribes a Chinese-language message without user language opt-in.

Low
Category
Output Handling
Confidence
92% confidence
Finding

The criterion requires the user-facing empty-list hint to be 「当前区域(x)无 CBR 备份」, fixing that output in Chinese. No language choice or justification for this language requirement is provided.

Content

No source excerpt is available for this finding.