Security audit

tradealpha实时新闻

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed TradeAlpha news lookup skill, but users should treat the required chat-provided token as sensitive.

Install only if you are comfortable pasting a TradeAlpha token into chat and having it sent to the listed TradeAlpha API domains. Prefer a limited-scope or easily revocable token, verify the TradeAlpha site yourself, and rotate the token or start a fresh conversation if you no longer want session reuse.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (9)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The examples explicitly instruct users to paste a live TradeAlphaToken directly into chat, but they do not warn that chat transcripts may be logged, retained, reviewed, or exposed to other tools and operators. Collecting bearer-style credentials through a conversational channel increases the chance of accidental disclosure and token theft.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill explicitly instructs users to paste a live API token into chat, which exposes a credential in a conversational channel without any warning, masking, or safer authentication flow. Even though the document says not to write the token to files or env vars, the token is still being collected and transmitted through the chat context, creating unnecessary secret exposure risk to logs, transcripts, downstream tooling, or later prompt leakage.

Ssd 3

Medium

Confidence: 97% confidence
Finding: The skill explicitly directs users to paste authentication tokens into chat and then reuse them across the session. Chat channels are often logged, retained, visible to platform operators, or accessible to downstream tooling, so collecting secrets this way increases the risk of credential leakage and unauthorized API use.

Ssd 3

Medium

Confidence: 98% confidence
Finding: The workflow makes token submission in chat a mandatory prerequisite before any request can proceed. This normalizes unsafe secret handling and encourages users to disclose credentials in a medium that may be logged or exposed beyond the immediate interaction.

Ssd 3

Medium

Confidence: 96% confidence
Finding: The error-handling path instructs users to resend the latest token in chat when authentication fails or expires. Repeatedly soliciting credentials in conversation compounds exposure risk, creates more copies in logs/history, and increases the chance of accidental disclosure or misuse.

Ssd 3

High

Confidence: 98% confidence
Finding: The skill instructs the assistant to collect an access token via chat and then use it for API calls, normalizing unsafe secret handling. If conversation content is logged, surfaced in analytics, or later revealed in prompt/context leakage, an attacker could reuse the token to access the user's TradeAlpha data and account entitlements.

Ssd 3

High

Confidence: 97% confidence
Finding: Telling the assistant to keep reusing a previously supplied token from conversation extends the lifetime and exposure of a sensitive credential inside the chat context. The more turns a secret remains in active memory or transcript history, the greater the chance of inadvertent disclosure, misuse by downstream tooling, or replay if the session is compromised.

Ssd 3

High

Confidence: 97% confidence
Finding: The token-refresh flow repeats the same unsafe pattern by asking the user to send a fresh token in chat, reinforcing credential sharing through an insecure channel. This increases the likelihood of repeated secret exposure and trains users into unsafe authentication behavior that could be exploited by phishing or transcript leakage.

Ssd 3

Medium

Confidence: 95% confidence
Finding: The instruction to remember and reuse a user-provided token for the rest of the conversation creates a secret-retention vulnerability in natural-language context. Persisting credentials in session memory increases the chance of accidental disclosure in later turns, prompt injection-induced exfiltration, or unintended reuse beyond the user's immediate intent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.