Back to skill

Security audit

Param Eval

Security checks for vulnerabilities and agentic risk

Overview

This is a simple Chinese procurement-parameter evaluation skill with no executable code, persistence, credential handling, or hidden behavior.

Before installing, confirm you want a Chinese-language workflow for tender/procurement parameter response tables and that any product documents placed under the referenced product directories are current and approved for use in generated bid materials.

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
  • 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 (5)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The description, triggers, workflow instructions, output template, and examples are all specified exclusively in Chinese, and the skill directs fixed Chinese prompts such as asking the user which product to use. There is no indication that the user may choose another language or that the Chinese-only behavior is a justified locale-specific requirement.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrase "需求响应" is broad enough to match many ordinary business or document-writing requests, which can cause the skill to activate outside its intended tender-parameter-evaluation context. Unintended activation can override user intent, pull in product-specific evaluation workflows, and increase the chance of irrelevant or misleading procurement output.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The phrase "技术评估" is highly generic and can apply to a wide range of unrelated requests, so the skill may trigger on routine analysis tasks that are not about bid/procurement parameter comparison. This broad matching creates scope confusion and may steer users into a constrained workflow that assumes specific product catalogs and response-table formatting.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger "产品对标" is ambiguous because it could refer to marketing comparison, internal benchmarking, feature comparison, or procurement evaluation. Without context limits, the skill may activate for unrelated comparison tasks and then inappropriately constrain the answer to the stored bid-evaluation framework and product directories.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The activation rule says the skill should activate when the user mentions any listed keyword, but it lacks scope guards, disambiguation logic, and exclusion conditions. This makes accidental invocation likely, especially because several listed terms are broad; in context, that can cause unintended file reads, product-specific assumptions, and low-quality or misleading outputs.

Content

No source excerpt is available for this finding.