ClawHub
Back to skill

Security audit

Torch Market

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Solana DeFi skill whose trading, vault, lending, and short-selling powers are high risk but coherent with its stated purpose.

Install only if you intend to let the agent build or submit Solana DeFi transactions. Use read-only mode or a fresh disposable controller key, never a vault authority key or seed phrase, set explicit spend limits, and expect token lookups to contact Solana RPC plus some third-party enrichment services.

SkillSpector

By NVIDIA
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 (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The skill markets itself as a narrowly scoped margin-market interface with 'no oracles, no stored baselines, no keepers,' but the documented functionality is materially broader and includes vault authority operations, token creation, protocol-maintenance cranks, rewards handling, token reclamation, and references to external enrichment/price sources. This kind of scope mismatch is dangerous because users or agent frameworks may grant trust based on the simplified description while the skill can drive higher-risk financial and administrative actions than expected.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The agent-facing section gives operational trading instructions such as creating a vault, depositing SOL, and calling buy/confirm flows, but it does not prominently warn that these actions can directly spend user-controlled assets and expose them to trading loss. In an autonomous-agent context, omission of an explicit funds-at-risk warning can cause unsafe use, over-trust in the tool, and accidental execution of real-money trades based on the reassuring audit language elsewhere in the document.

External Transmission

Medium
Category
Data Exfiltration
Content
// Fetch SOL price
    let solPriceUsd;
    try {
        const res = await fetch('https://api.coingecko.com/api/v3/simple/price?ids=solana&vs_currencies=usd');
        const data = (await res.json());
        solPriceUsd = data?.solana?.usd;
    }
Confidence
93% confidence
Finding
https://api.coingecko.com/

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.