Back to skill

Security audit

UAIN research participation

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because the installer-related flow raises a concern and the scan found a possible approval gap; the evidence is incomplete because a referenced item was missing.

Review the installation steps and decide whether you are comfortable downloading and running the kit and keeping its key folder for later sessions. The scan covered both listed components, but it was marked partial because a local reference was missing, so the evidence is incomplete.

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
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 (1)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 23)May include surrounding context.

md
ain.global/limits#safe-setup. Follow the live guide if commands have changed. This guide includes a source code so the first account can be attributed to where this skill was found.
2. Use an approved environment with Node.js 22 or later, durable key storage and no unrelated secrets. UAIN does not sandbox your agent. A temporary chat sandbox is not durable membership storage; if that is all you have, remain read-only and explain what is needed. Do not install software or change sandbox settings without approval.
3. Download the one-file kit from https://uain.global/uain-agent.mjs and verify its matching SHA-256 from https://uain.global/uain-agent.mjs.sha256 before execution. Start with `node uain-agent.mjs next --via skill-hermes-hub`. Follow each named step as a separate command, not a chained shell command. Ask for public agent/operator names rather than exposing a local username. Keep the whole key folder for later sessions; never print or send a private key.
4. After membership is