Back to skill

Security audit

私募尽调

Security checks for vulnerabilities and agentic risk

Overview

The skill’s purpose is coherent, but it asks users to install or update and then run mutable third-party code using local Cue credentials.

Review the external cue-skills repository before running this skill, prefer a pinned commit or trusted packaged runner, and run it in a least-privilege environment. Be aware it may use your Cue API key and spend credits after confirmation.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The "何时用" section describes the skill as applicable to broad domains like enterprise due diligence, funds/institutions, and disclosure/regulation, but it does not specify concrete trigger phrases, boundaries, or cases where the skill should not activate. For a markdown skill description, this ambiguity can cause over-selection for loosely related user requests in common finance or compliance contexts.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill instructs the agent/user to git pull or clone a remote repository and then execute a Python runner from that repository before performing the task. This creates a software supply-chain risk: a compromised GitHub/Gitee repo, account takeover, malicious dependency change, or unexpected code update could lead to arbitrary code execution on the host with the user's privileges.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill tells users to clone/update an external repository and run its script, but it omits a meaningful warning that this executes externally sourced code and may change over time. In a skill context, that omission is dangerous because it normalizes unsafe execution of unreviewed remote content and increases the chance of supply-chain compromise.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
75% confidence
Finding

All user-facing instructions and operational guidance in the skill are presented only in Chinese, and the file does not indicate that the language is optional or region-justified. Under the policy, forcing a specific language without opt-in can be a natural-language locale violation.

Content

No source excerpt is available for this finding.