Back to skill

Security audit

Configure File Write And Recovery

Security checks for vulnerabilities and agentic risk

Overview

The skill is not malicious, but it gives broad guidance and a bundled script for overwriting gateway configuration without enough safeguards.

Install only if you intentionally want a Chinese-language skill that helps modify persistent gateway configuration. Before using it, require explicit confirmation, back up the existing config, validate model IDs and syntax, use a real scoped path, and have a rollback plan before restarting the gateway.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (10)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill instructs the agent to write configuration files and invoke shell-based execution paths, but the manifest does not declare any explicit tool scope or permissions. This creates a capability-transparency gap: a caller or platform reviewer cannot easily tell that the skill may modify files or use shell execution, increasing the chance of unsafe invocation or over-privileged operation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger conditions are very broad and match common topics like configuration writing, restart behavior, and model configuration. That increases the likelihood the skill activates in routine conversations and offers risky guidance for file overwrites and gateway restarts in contexts where the user did not explicitly request such operational changes.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill provides step-by-step instructions to overwrite configuration files and reboot a gateway without prominently warning about backup requirements, rollback steps, downtime, or the risk of configuration loss. In operational environments, this can directly cause service disruption, broken boot-time validation, or persistent misconfiguration if users follow the instructions blindly.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The evaluation prompt describes very common troubleshooting symptoms like configuration loss after reboot without sufficiently constraining product, environment, or required context. This can cause the skill to activate in unrelated conversations and steer users toward file-writing or configuration-modification guidance in contexts where it is inappropriate or risky.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Phrases such as '这是怎么回事' and '有没有什么方法可以解决' are generic help-seeking language that appear in many benign troubleshooting requests. When paired with only loosely scoped config-write behavior, they increase accidental invocation risk and may lead the agent to provide sensitive operational instructions outside the intended scenario.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script unconditionally opens the target path in write mode, which truncates and replaces the existing configuration file without backup, validation, or confirmation. In a skill specifically about persistent gateway configuration, this is risky because a wrong path, malformed content, or unintended execution can break service configuration and cause loss of previous settings across restarts.

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/write_config.py (reported line 22)May include surrounding context.

python
with open('/path/to/config/file', 'w') as f:
    f.write(config_content)

subprocess.run(['exec', 'cat', '/path/to/config/file'])

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill content is presented entirely in Chinese, including the core description and operating instructions, with no indication that the user opted into that language or that the skill is intentionally limited to a Chinese-speaking context. This can violate language or locale policy when users are not given a choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

Natural-language policy review applies to all file types. The eval content is entirely in Chinese and does not indicate multilingual support, user opt-in, or that the skill is intentionally region- or locale-specific, which can amount to forcing a language without explicit choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The only natural-language guidance in the file is a Chinese comment, which imposes a specific language context without any indication of user choice or locale justification. Under the policy, forcing a language without opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.