ClawHub
Back to skill

Security audit

Translation Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent translation skill with a small file-writing caveat when users give it a file path.

Before using this skill on a file path, expect the agent to read that file and create a translated copy with a suffix. Avoid giving paths to confidential files unless you are comfortable having their contents translated in the agent environment.

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 (4)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is presented as a translation assistant, but it instructs the agent to read a file from a user-provided path and write a new translated file automatically. That expands the skill from text transformation into filesystem access and modification, creating risk of unintended file reads, overwrites, path traversal abuse, or exfiltration of sensitive local content if an agent follows the instruction without strict safeguards.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The markdown explicitly tells the agent to read a file and write a translated output file, but it does not warn the user that this would modify the filesystem or require consent. In an agent environment, silent file creation or modification is dangerous because users may believe they are only requesting translation while the skill performs side effects on local or connected storage.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The instructions say the default is zh-CN → en for business/tech and en → zh-CN for learning, which imposes a language-direction choice before the user explicitly selects it. SQP-3 covers language or locale policy violations when a skill forces a specific language without user opt-in, and this behavior is only partially mitigated because confirmation is required only for reverse or other cases.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · SKILL.md (reported line 93)May include surrounding context.

md
- Keep standard legal boilerplate terms:
  - 本协议 → this Agreement (capitalized)
  - 一方 → a Party / either Party
  - 包括但不限于 → including but not limited to
  - 自生效之日起 → as of the Effective Date
  - 具有法律约束力 → legally binding

Static analysis

No suspicious patterns detected.