ClawHub
Back to skill

Security audit

Skill Evolver Pro - 技能进化器专业版

Security checks for vulnerabilities and agentic risk

Overview

The skill is a local analytics and evolution helper with disclosed disk logging, but users should avoid feeding it sensitive raw logs or feedback.

Install only if you are comfortable with local storage of skill telemetry and evolution history. Use a dedicated data directory, keep its permissions restricted, and redact secrets, personal data, prompts, customer content, and detailed stack traces before passing logs or feedback into the pipeline.

SkillSpector

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly promotes persistent storage of evolution logs, version history, state, usage statistics, error logs, and user feedback, but provides no warning, minimization guidance, retention policy, or access-control expectations for potentially sensitive data written to disk. In a self-evolving analytics skill, these artifacts can contain private prompts, operational metadata, and user feedback that may be exposed through local compromise, backups, shared home directories, or improper file permissions.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
The pipeline persists an evolution log containing collected stage data, and earlier stages ingest potentially sensitive inputs such as error logs, user feedback, and performance metrics. Writing this data to storage without minimization, redaction, retention controls, or any visible consent/disclosure mechanism increases the risk of inadvertent exposure of sensitive operational or user-derived information.

Static analysis

No suspicious patterns detected.