ClawHub
Back to skill

Security audit

Mpp Mobula

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed pay-per-call crypto data client that stores a low-balance hot wallet and automatically spends small USDC.e amounts as its core function.

Install only if you want an agent to make paid Mobula API calls from a funded Tempo wallet. Keep only a small balance, monitor usage, review the viem dependency/source repo, and avoid running it on shared or untrusted machines because repeated calls can still spend the wallet down in small increments.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (5)

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The skill metadata says it provides crypto prices, wallet positions, and market data, but the code also exposes wallet activity/transaction history retrieval. That creates a scope mismatch that can lead to unexpected collection or exposure of more sensitive financial behavior data than users or reviewers were told about, especially in a Telegram-linked per-user context.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The README states that the skill will automatically sign and broadcast a USDC.e payment transaction for each API call, but it does not prominently warn users that invoking the skill spends real funds. In an agent context, this is risky because repeated or untrusted tool use can silently drain a wallet through many small charges, especially when calls are triggered programmatically on behalf of users.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill promotes automatic on-chain payments and wallet persistence with language like 'every subsequent call is automatic' and 'no manual signing' but does not present a strong, explicit warning that future API calls spend real funds and rely on stored secret material. In an agent context, this is dangerous because routine prompts could trigger repeated micropayments or wallet use without the user realizing they are authorizing financial transactions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
This function accepts a raw private key and can immediately sign and broadcast a payment transaction in response to a 402 challenge, with no explicit user confirmation, consent checkpoint, or spending-policy enforcement beyond a per-call cap. In a pay-per-call skill, that means any code path reaching tempoFetch can spend user funds automatically and repeatedly, which is especially risky for agentic or background execution where users may not realize a blockchain payment is occurring.

Session Persistence

Medium
Category
Rogue Agent
Content
cd mpp-skill
bun install

# 1. Create a Tempo wallet (saved at ~/.mpp-skill/wallet.json, chmod 600)
bun run start wallet-create

# 2. Fund it with USDC.e via the bridge link printed by step 1
Confidence
81% confidence
Finding
Create a Tempo wallet (saved at ~/.mpp-skill/wallet.json, chmod 600) bun run start wallet-create # 2. Fund it with USDC.e via the bridge link printed by step 1 # https://relay.link/bridge/tempo?to

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.