Security audit

Convene

Security checks across malware telemetry and agentic risk

Overview

The skill’s core TradingView-style integration appears coherent, but its credential setup asks for an API key through chat and persists it locally, which users should review before installing.

Install only if you are comfortable configuring the required API key. Prefer adding the key yourself in a local secret store or environment file, avoid pasting it into chat, use the narrowest-scoped/revocable key available, and rotate it if it was ever shared in a conversation.

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 (2)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill instructs the agent to have users provide an API key and then persist it locally, including via chat on remote channels, without an explicit warning about the risks of transmitting secrets in conversation logs. This creates a credential-handling workflow that can expose secrets through chat retention, agent logs, or unintended operator access.

Ssd 3

Medium

Confidence: 97% confidence
Finding: This section explicitly creates a natural-language secret collection flow by telling the agent to ask for a TradingView API key and save it to a local .env file, including over remote channels. Even if intended for convenience, prompting users to disclose credentials in chat increases the chance of leakage through logs, transcripts, or mishandled deletion semantics.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.