ClawHub
Back to skill

Security audit

Auto Research Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed autonomous training-experiment workflow that edits and runs a local demo training file, with no evidence of hidden access, exfiltration, or persistence.

Install only if you want an agent to iteratively edit train.py and run local training experiments. Use version control or a disposable project copy, and be mindful of CPU/GPU time before starting the research loop.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly directs an agent to inspect, modify, and then run local code as part of its normal workflow, but it provides no user-facing warning, confirmation step, sandboxing guidance, or safety boundary around code execution and file modification. In an autonomous research context, this is especially risky because repeated iterative edits and executions can amplify accidental damage, execute unsafe repository code, consume significant resources, or alter project state without informed user consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The quick-start section tells the agent to run python train.py directly, again without any visible warning about code execution, resource usage, external effects, or the possibility that train.py may contain unsafe logic. Because this skill is designed for autonomous iteration, normalizing immediate execution increases the chance that an agent will launch unreviewed training code repeatedly, causing system impact, unexpected network or file activity, and unintended cost or stability issues.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This code file contains natural-language content in Chinese in the module docstring and comments, which effectively imposes a language choice on readers or maintainers without any opt-in or explanation. Under the policy, forcing a specific language or locale without user choice is a natural-language policy violation.

Content

No source excerpt is available for this finding.