Back to skill

Security audit

贷后监测

Security checks for vulnerabilities and agentic risk

Overview

The skill’s loan-monitoring purpose is coherent, but it asks users to fetch and run mutable external code that uses a local Cue API key, so it should be reviewed before installation.

Install only if you trust cuecue.cn and the sensedeal/cue-skills repository. Prefer reviewing or pinning the runner before execution, and understand that your query plus the Cue API key from ~/.cue/config.json will be used by the external Cue workflow and may consume credits.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The entire skill description and usage instructions are written only in Chinese, including the prescribed user-facing confirmation text at L44. This imposes a specific language/locale on users without any opt-in or alternative, which matches the policy-violation criterion for language or locale restrictions.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs the operator to clone or update and then trust code from external Git repositories before performing the advertised task. That introduces unnecessary supply-chain risk: a compromised upstream repo, mirror, or account could deliver malicious code that runs locally under the user's account, which is materially broader than a documentation-only post-loan monitoring skill.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill directs the user to run a local Python script that automatically reads API credentials from ~/.cue/config.json and sends user-supplied queries to an external service. This creates a credential-handling and code-execution boundary the manifest does not make explicit, and if the runner is malicious or modified it could exfiltrate local secrets or misuse the API key.

Content

No source excerpt is available for this finding.