Security audit

A股每日复盘模板

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple A-share review-template generator that writes user-requested Markdown files and does not show hidden, network, credential, persistence, or destructive behavior.

Install only if you want Chinese-language A-share trading review templates. Use the default output directory or choose --output carefully, because the script writes directly to that path and may overwrite an existing Markdown file.

A.I.G

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

SkillSpector

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 95% confidence
Finding: The skill instructs the agent to write Markdown files to user-controlled paths such as `~/a-share-reviews/...` or an arbitrary `--output` location, but it does not declare any tool scope or permission constraints. That creates an authorization gap where file-write capability may be available without explicit limitation, increasing the risk of unintended overwrites or writes to sensitive locations if the command is invoked with crafted output paths.

Natural-Language Policy Violations

Medium

Confidence: 96% confidence
Finding: This Python skill’s user-facing description and generated templates are all Chinese-language only, and the CLI help text later follows the same pattern. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified.

Natural-Language Policy Violations

Low

Confidence: 91% confidence
Finding: This markdown file applies to the natural-language policy category, which includes language or locale policy violations. The README presents all user-facing instructions exclusively in Chinese and does not provide an opt-in, alternative language, or justification that the skill is region- or language-specific.

Intent-Code Divergence

Low

Confidence: 99% confidence
Finding: The top-level documentation states '支持6种复盘风格', but TEMPLATES includes seven keys: 情绪、龙头、趋势、数据、个人、游资、完整. This is an active contradiction between the file documentation and the implemented functionality.

Description-Behavior Mismatch

Low

Confidence: 98% confidence
Finding: The manifest says the skill supports 'auto-filled data fields' and mentions 7 styles. In the implementation, template generation only fills {date} and {time} via string formatting, while all market data fields remain blank placeholders like '___'; the available styles are simply static template variants. This creates a mismatch between the advertised level of data population and the actual behavior.

Static analysis

No suspicious patterns detected.