Back to skill

Security audit

huawei-cloud-nickname-generate

Security checks for vulnerabilities and agentic risk

Overview

This skill generates nicknames locally with proportionate file access and no credential, cloud, or network use.

No cloud credentials or network access are needed. Choose output paths carefully because existing files are overwritten. Incomplete custom pools can relax style matching with a notice; generated names are suggestions without availability or trademark checks.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The description explicitly says a selected style filters every slot to morphemes carrying that style. In _generate_one, when a slot has no matching morphemes, the code retains the original pool, sets strict=False, and continues generating. It reports the relaxation afterward, but this still contradicts the declared filtering guarantee. Otherwise, the code's primary purpose and resource use match the description; no network, cloud, subprocess, or unrelated behavior is present.

Content

No source excerpt is available for this finding.

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

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 · tests/test_generate_nickname.py (reported line 222)May include surrounding context.

python
class CliTests(unittest.TestCase):
    def _run(self, *args):
        return subprocess.run(
            [sys.executable, SCRIPT_PATH, *args],
            capture_output=True,
            text=True,

The trigger “取名字” (“choose a name”) is broader than nickname generation and can activate the skill for unrelated naming requests.

Low
Category
Trigger Abuse
Confidence
87% confidence
Finding

“取名字” can refer to naming a child, business, product, or project, rather than generating a nickname. Including it as a standalone trigger without a nickname-specific qualification or exclusion can cause unintended invocation.

Content

No source excerpt is available for this finding.

The requested style is silently relaxed during generation rather than enforced for every morpheme slot.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The manifest promises that a selected style filters every slot to morphemes carrying that style. At L177–L187, when no matching morphemes exist for a slot, the code retains the unfiltered pool and selects from it, allowing returned names to violate an explicitly requested style. Although the result records this fallback and the renderers disclose it, that behavior still contradicts the advertised filtering guarantee.

Content

No source excerpt is available for this finding.