股票数据API

Security checks across malware telemetry and agentic risk

Overview

This is a stock-market data lookup skill that uses an API key to query data.diemeng.chat, with no evidence of hidden collection, destructive behavior, or privilege escalation.

Install only if you trust data.diemeng.chat with your API key and the market-data queries you ask the agent to run. Configure the key through OpenClaw's secret or environment mechanism, avoid putting real keys in source code, and use narrow symbols, date ranges, and limits to control quota and data exposure.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Description-Behavior Mismatch

Medium

Confidence: 85% confidence
Finding: The skill includes an endpoint for reading snapshot push-history records described as Redis/Kafka push records, which is materially different from ordinary end-user stock queries and is not disclosed in the skill description. Access to internal message-history style data can expose operational or infrastructure-derived records that users would not reasonably expect this skill to access.

Context-Inappropriate Capability

Medium

Confidence: 90% confidence
Finding: A stock-query skill does not need access to Redis/Kafka push-channel history to fulfill its stated purpose, so this endpoint represents unjustified capability expansion. Even if the remote API enforces authorization, exposing this function in the skill increases the chance of collecting or revealing backend operational data outside the expected user task.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal