Security audit

股智Alpha

Security checks across malware telemetry and agentic risk

Overview

This stock-analysis skill is mostly purpose-aligned, but it deserves review because it can trigger broadly and uses unsecured third-party market data for financial recommendations.

Install only if you are comfortable with third-party market-data lookups and treat results as informational, not investment advice. Verify important quotes and recommendations against trusted sources, use an isolated Python environment, and be aware that the documented scan scope and fund-flow source may not match the default script behavior.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Confidence: 83% confidence
Finding: The trigger phrases are broad and overlap with ordinary financial conversation such as '今天买什么', '股票推荐', and '现在大盘怎么样'. This can cause accidental invocation in contexts where the user did not intend to run a data-fetching, recommendation-generating skill, leading to unnecessary network activity and potentially sensitive portfolio analysis without clear consent.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The code fetches stock market data over plain HTTP from a third-party endpoint, so requests and responses can be intercepted or modified by any network attacker between the agent and the provider. In this skill context, tampered market data could directly influence stock screening and trading decisions, making integrity loss more important than the apparent sensitivity of the query itself.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: This code performs additional unauthenticated third-party network calls to Tencent over HTTP without informing users or protecting transport security. Besides privacy concerns, an attacker able to tamper with the plaintext response could inject false intraday pricing data that propagates into the skill's analysis results.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The Sina API call sends requests over plain HTTP to an external market-data service with no user-facing disclosure. Because this skill is an A-share stock selection system, manipulated or spoofed response data could corrupt factor calculations and recommendations, potentially causing financial harm to users relying on the output.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The code sends stock symbols to a third-party Sina endpoint over plain HTTP, which exposes requests and responses to interception or tampering by any network attacker on the path. In this finance-related skill, even though the queried symbols are not highly sensitive secrets, insecure transport can leak user interest patterns and allow manipulated market data to be injected into downstream analysis or decision-making.

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.