ClawHub
Back to skill

Security audit

OATDA Compare Models

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward model-comparison helper that uses a disclosed OATDA API key and sends user prompts to OATDA for multi-model results.

Install only if you are comfortable sending prompts through OATDA and its connected model providers. Treat prompts as external API data, keep the OATDA API key private, and avoid using this skill for secrets or confidential content unless your OATDA account and data-handling expectations allow it.

SkillSpector

Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Credential Access

High
Category
Privilege Escalation
Content
"openclaw":
      {
        "emoji": "⚔️",
        "requires": { "bins": ["curl", "jq"], "env": ["OATDA_API_KEY"], "config": ["~/.oatda/credentials.json"] },
        "primaryEnv": "OATDA_API_KEY",
      },
  }
Confidence
88% confidence
Finding
The skill instructs the agent to read an API key from ~/.oatda/credentials.json and export it into the shell environment for subsequent curl calls. Although this is framed as normal authentication handling and the skill warns not to print the full key, automatic credential sourcing from a local file expands secret exposure risk if command construction, logging, subprocess inspection, or later shell steps leak environment variables.

Static analysis

No suspicious patterns detected.