Back to skill

Security audit

euthyna

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because it reports two medium severity concerns about running project tests and exploit demonstrations without requiring safe containment; the available scan is partly incomplete.

Consider the reported execution risks before installing. If you use the skill, review its verification guidance and ensure project tests or exploit demonstrations run only in an isolated, disposable environment. Skillspector marked its analysis partial because a local path-like reference could not be resolved; the AIG scan completed with no findings, but that does not resolve the incomplete coverage.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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)

The coverage workflow instructs the agent to execute audited project code without warning about its potential effects on the host.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The coverage workflow directs the agent to run audited-project tests without first requiring isolation or establishing that the project is trusted. In a security-audit context, attacker-controlled tests, initialization code, or build hooks could execute with the agent's host permissions, exposing credentials or modifying user files; the option to mark coverage unevaluated does not establish a safe execution boundary.

Content

No source excerpt is available for this finding.

The workflow requires executing exploit PoCs without warning about effects on user data or system integrity.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The verification rules require actual PoC execution and captured effects but do not require containment or authorization for potentially destructive behavior. Because these PoCs intentionally exercise vulnerabilities, running them against the host or a live target could compromise credentials, corrupt data, or disrupt services; restrictions on which subagent may write do not contain executed code.

Content

No source excerpt is available for this finding.