Back to skill

Security audit

资料到成果交付助手

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed file-delivery workflow with optional AI-HIVE/MCP use and a local validator, with no hidden persistence or credential collection found.

Install only if you want a Chinese-language workflow for turning provided materials into concrete deliverable files. Before using AI-HIVE or any external tool, confirm exactly which files may be uploaded, the budget, and any external actions such as sending, payment, deletion, or publishing.

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 (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The declared description presents a production-oriented assistant that transforms user inputs into delivery outputs. The supplied code does something materially different: it validates the structure and provenance of an already-produced result file and optionally checks local artifact hashes under an allowed root. It contains no logic for creating deliverables, mapping sources, generating checklists, orchestrating tools, or using AI/MCP. Its primary purpose is offline compliance/validation, not成果生成或交付执行. Therefore the description does not accurately represent the code's actual behavior.

Content

No source excerpt is available for this finding.

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.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation language is broad enough to match many ordinary assistant requests, increasing the chance the skill is invoked outside its intended context. Over-broad routing can cause unnecessary exposure of user files, unintended workflow transitions, or inappropriate attempts to connect to external tooling when a simpler response would suffice.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

Ambiguous scope makes it unclear which user requests should invoke the skill, which can lead to overreach into tasks that do not require document-processing or external-tool workflows. In a file-handling assistant, ambiguity increases the risk of mishandling sensitive inputs or initiating tool-dependent processes without clear user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The default activation prompt is very broad and will trigger the skill for loosely related requests involving materials, delivery, or document tasks without strong gating conditions. In an agent environment, this can cause over-activation, unnecessary access to user files or connected tools, and increased exposure to prompt injection or unintended task execution through attached materials.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The entire skill file is written as if operation and outputs will be conducted in Chinese, with no statement that the user may choose another language or locale. Under the policy rules, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is documented and justified.

Content

No source excerpt is available for this finding.