Back to skill

Security audit

Self Improvement Daemon

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to describe a persistent self-improvement daemon with broad triggers and unclear behavior, but the reviewed artifact does not show confirmed malicious code or exfiltration.

Review this carefully before installing. It may be harmless documentation, but the skill claims daemon-style self-improvement behavior without enough detail about what will run, what it can modify, or how users approve and stop it. Do not enable broad auto-triggering for terms like "improve" unless the actual daemon code is available and reviewed.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger phrases are broad everyday terms like '自我改进', '升级系统', and 'improve', which can cause accidental invocation during normal conversation. In the context of a persistent self-improvement daemon, unintended activation could launch state-changing behavior without clear user intent, increasing the risk of unauthorized system modifications or persistence.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation presents a daemon intended for continuous self-improvement but does not warn users that it may run persistently or alter system state. Without explicit warnings and consent boundaries, users may invoke a long-running process without understanding operational, security, or stability consequences.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The manifest description, headings, and trigger phrases are entirely in Chinese except for one generic English trigger, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-only context. This can constitute a language-policy issue when the skill imposes a locale without opt-in or documented justification.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill description claims a daemon that continuously learns to optimize the system, which implies code or documentation about automated improvement behavior. Instead, most of the file is populated with manually accumulated Bilibili video lists about unrelated 'daemon' or 'self-improvement' keyword matches, which does not align with the claimed operational purpose.

Content

No source excerpt is available for this finding.