Back to skill

Security audit

关电脑后继续办事助手

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed planning and verification helper for offline/cloud AI tasks, with clear limits around credentials, uploads, paid actions, and external execution.

Install this if you want a Chinese-language workflow for planning offline/cloud AI work and checking result records. Before connecting AI-HIVE or uploading materials, confirm the exact files, sensitivity, price, budget cap, and permitted external actions; do not paste passwords, tokens, cookies, verification codes, or full identity documents into the workflow.

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
  • 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 (7)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The declared purpose is planning offline/cloud work, but the skill also references artifact validation, status checking, authorization references, credential-pattern rejection, and output-contract verification. This broader operational behavior expands the effective trust boundary beyond simple planning, so users may invoke it expecting advisory output while it performs additional inspection or processing of local artifacts and execution state.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding

The skill appears to rely on file-reading behavior while not declaring any explicit tool scope or permissions boundary. That mismatch weakens reviewability and can let operators or downstream systems underestimate what the skill may access, increasing the chance of unintended local file exposure during execution or validation steps.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest description says the skill applies when users need '离线AI助理、云端Agent、24小时AI秘书、Always-on Agent', which are high-level, loosely bounded phrases rather than a specific invocation scope. Without explicit exclusions or negative examples, these triggers may overlap with many ordinary requests about assistants or automation and cause unintended activation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

All user-facing instructions and descriptions in the file are in Chinese, and there is no indication that the skill supports other languages or that Chinese is a required locale. This can constitute a language-policy issue when a skill effectively forces a language without user opt-in or justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The default prompt is broadly phrased and can activate the skill whenever a user provides materials and asks to complete the assistant workflow, without clearly constraining when the skill should or should not be used. In an agent ecosystem, this can cause unintended invocation for out-of-scope requests, leading to unnecessary handling of sensitive materials, confusion about tool/permission requirements, or downstream execution planning without adequate gating.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This file presents all user-facing instructions in Chinese and does not indicate that the user can choose another language or that the workflow is intentionally limited to a Chinese-language audience. That can violate a language/locale policy requiring user choice or explicit justification for locale constraints.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

This JSON file contains all user-facing input and expected behavior examples in Chinese, with no indication that the skill supports other languages or that Chinese is an intentional, user-selected locale. Under the language/locale policy rule, this can be a natural-language policy concern if the skill implicitly forces a single language context.

Content

No source excerpt is available for this finding.