Security audit

云之家表单通用解析规则

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only YunZhijia form parsing skill, with no executable code or hidden behavior found.

Safe to install as reference documentation. Use it only with YunZhijia form data you intend to process, confirm before downloading attachments, and do not paste or store real YunZhijia access tokens in the skill files or shared prompts.

SkillSpector

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 (1)

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The README states that the skill will be automatically triggered when users broadly describe tasks like parsing YunZhijia approval push data or building approval request payloads. Because these trigger descriptions are generic and overlap with normal enterprise coding and data-processing requests, the skill may activate in contexts the user did not explicitly intend, causing unintended access to or transformation of sensitive workflow/form data.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: references/file-widget.md:71