Security audit

vedetta

Security checks for vulnerabilities and agentic risk

Overview

The skill openly buys market research, but its payment helper does not enforce the advertised prices before authorizing payments.

Use the free endpoints until payment checks enforce each route's price, an overall budget, and the expected network, USDC asset, and recipient before signing. If enabling payments, use a dedicated low-balance wallet. Do not rely on the written price warning alone. The absent Solana helper and external packages remain outside the verified artifact behavior.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md
Finding: Automatic payment client does not enforce advertised spending limits
Content: View full analysis
Impact

The client can authorize a payment above the expected $0.01-$0.09 route price, potentially spending the wallet's available USDC rather than the intended small fee.

SkillSpector

Vulnerability Patterns

Rogue AgentSelf-Modification, Session Persistence
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

The payment client automatically accepts x402 payments without enforcing the documented price ceiling.

High

Category: Excessive Agency
Confidence: 96% confidence
Finding: The example payment client delegates signing to the x402 wrapper without enforcing the documented per-endpoint prices or $0.09 maximum before authorization. A compromised or malicious endpoint could return an inflated payment offer and obtain an authorization exceeding the user's intended spend; the low-balance-wallet recommendation limits exposure but does not enforce the ceiling.
Content: No source excerpt is available for this finding.

Session Persistence

Medium