Back to skill

Security audit

huawei-cloud-eip-bandwidth-list

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence, mainly because installation runs a downloaded script and environment settings may redirect a subprocess to another program.

Review the installer before running it and verify it through a trusted source if possible. Consider whether the skill’s shell and file capabilities and environment overrides are acceptable in your setup, and use only with credentials and permissions you are comfortable exposing to the invoked program. The scan was partial: five local references were unresolved, 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
  • 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 (6)

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 installation instructions execute a downloaded shell script without warning about local system changes or remote-code execution.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The installation procedure executes a remotely downloaded, mutable shell script without verifying its integrity or reviewing its contents. HTTPS protects transport but not against compromise of the distribution server or installer, which could execute attacker-controlled code with the installing user's filesystem and credential access; the cloud operation allowlist does not constrain installation.

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_bandwidths.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
87% confidence
Finding

The inherited environment can replace the executable through EIP_BANDWIDTH_HCLOUD_BIN and prepend arbitrary arguments through EIP_BANDWIDTH_HCLOUD_ARGS, outside the operation guard. If a lower-trust launcher or environment configuration can influence these variables, an attacker can run another program with the agent's privileges and inherited cloud credentials, bypassing the read-only command construction without needing shell injection. Exploitation requires environment control; ordinary bandwidth query arguments alone do not provide this capability.

Content

Scanner excerpt · scripts/list_bandwidths.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,
        )

Default table headers and Markdown report text are fixed in Chinese without a language choice or documented locale requirement.

Low
Category
Output Handling
Confidence
93% confidence
Finding

The output specification requires Chinese headers, report titles, and empty-result text regardless of the user's language. No language opt-in or justification for restricting these outputs to Chinese is provided; JSON output changes the format rather than offering a locale choice.

Content

No source excerpt is available for this finding.

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

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The module documentation at L0010–L0012 says credentials are read by the CLI or SDK and that this script never reads secrets. In contrast, L0544–L0545 retrieve AK/SK values through _env_value (L0475–L0479), and L0561 passes them into BasicCredentials. Credential access is justified by the SDK fallback, but the explicit secret-handling assurance is inaccurate; this does not itself demonstrate credential disclosure.

Content

No source excerpt is available for this finding.