Security audit

蜡烛图分析

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a coherent stock charting tool, with expected network data fetching and chart file output but some usability and disclosure limitations.

Install only if you are comfortable with a Chinese-language stock analysis script that contacts akshare/Tencent for market data and writes chart PNG files to your Desktop using the stock code as the filename. Re-running the same stock code may replace the prior chart.

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

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: This Python file hard-codes Chinese-language descriptions and user-facing output, beginning with the module description on L002 and continuing throughout titles and printed messages. The policy for natural-language violations applies to all file types, and there is no opt-in or configurable language selection for users who may require another locale.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The skill has a side effect beyond pure analysis: it writes an output image directly to the user's Desktop. While saving a generated chart is related to the stated functionality, doing so automatically to a user-visible location without consent or configurability is an unsafe behavior because it can overwrite files, expose sensitive stock activity on a shared machine, or violate least-surprise expectations.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The code writes to a hardcoded path under ~/Desktop, which assumes a specific environment and performs filesystem writes without user choice. This expands the skill's privileges unnecessarily and can disclose user activity, fail unpredictably across systems, or overwrite artifacts in a sensitive user-facing directory.

Missing User Warnings

Low

Confidence: 89% confidence
Finding: The skill explicitly states it uses akshare/Tencent to fetch stock market data, but the usage instructions do not warn users that running the script will initiate outbound network access. This is a real transparency and trust issue: users may execute the tool assuming it is local-only, which can violate user expectations, policy constraints, or restricted-network environments.

Static analysis

No suspicious patterns detected.