ClawHub
Back to skill

Security audit

Which LLM? Deterministic model selection for agents

Security checks for vulnerabilities and agentic risk

Overview

This instruction-only skill is coherently scoped to a paid Which-LLM API flow, but users should handle wallet payments with explicit approval and address verification.

Install only if you are comfortable with an external API receiving model-selection goals and outcome data, and configure wallet/payment handling so every transfer requires explicit user approval, verified chain, verified recipient, exact amount, and matching payment reference.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly requires a crypto wallet for paid requests and describes a 402/payment-proof flow, but it does not warn that blockchain transfers are real-value, potentially irreversible, and vulnerable to mispayment if chain, recipient, or amount are wrong. In a high-sensitivity skill that may induce autonomous or semi-autonomous payment behavior, this omission materially increases the risk of financial loss and unsafe operator assumptions.

Content

No source excerpt is available for this finding.

Unrestricted Tool Access

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill grants unrestricted tool access without appropriate constraints. An agent with unfettered tool access can perform arbitrary actions including file modification, network requests, and code execution.

Content

Scanner excerpt · SKILL.md (reported line 66)May include surrounding context.

md
## What this skill does not do

- It does not call an LLM directly
- It does not execute arbitrary code from your prompt
- It does not require unrelated files or system access beyond API usage and payment handling

## Runtime Scope

Static analysis

No suspicious patterns detected.