Back to skill

Security audit

Model Router

Security checks for vulnerabilities and agentic risk

Overview

This is a simple model recommendation helper with minor disclosure and usability issues, but no evidence of malicious behavior.

Install only if you are comfortable with a Chinese-language model routing helper. Before following its configuration example, back up your OpenClaw config and validate the JSON; also note that the Python module will read /home/skyswind/.openclaw/openclaw.json if that file exists.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (8)

Undeclared Tool Scope

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.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Most user-facing instructions and descriptions in the skill are written exclusively in Chinese, and the file does not indicate that Chinese is optional or required for a region-specific purpose. This can violate language/locale policy when a skill effectively forces a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The module docstring and later user-facing strings are written only in Chinese, indicating the skill is designed to operate in a single language. The policy requires avoiding forced language or locale constraints unless the skill offers user opt-in or clearly documents a justified region-specific limitation, which this file does not do.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest promises model selection based on task type, content length, and quality, but the code also consults local agent configuration for model defaults. This hidden behavior creates a trust-boundary mismatch: consumers may invoke the skill expecting pure routing logic, while it also inspects local state, which can expose configuration details and produce non-obvious behavior.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill reads a user-specific OpenClaw configuration file and exposes get_agent_model, which reveals per-agent configured model choices unrelated to the manifest’s stated routing function. In a skill context, unnecessary access to local user configuration expands the data-access surface and can leak environment-specific information to other components or prompts that invoke this helper.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The explanatory strings returned to users, such as model descriptions and task-fit labels, are all emitted in Chinese. Because the file does not offer alternative locales or an explicit opt-in, it violates the language-choice policy for natural-language behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The command-line help text shown to users is entirely in Chinese, and there is no option to select another language. This creates a natural-language policy issue because the skill enforces a specific language for user interaction without documented opt-in or locale justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The documentation instructs users to directly edit a live OpenClaw configuration file under the user's home directory without cautionary guidance. This can lead to accidental service misconfiguration, broken agent routing, or loss of a known-good setup, especially if users overwrite production settings without backup.

Content

No source excerpt is available for this finding.