ClawHub
Back to skill

Security audit

成功的项目管理

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed project-memory and documentation helper with opt-in setup and history import behavior, not hidden or destructive behavior.

Before installing, be comfortable with a skill that can maintain project memory files and documentation. Use explicit requests for setup, adaptation, or history import, and avoid asking it to store secrets or unrelated private chat content.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill's stated purpose is project memory and documentation maintenance, but its referenced behavior includes inspecting history sources and inventorying session or export metadata. That mismatch is dangerous because users may invoke the skill expecting routine documentation help while it also enables broader discovery of local chat/history artifacts, which can expose sensitive data or expand access beyond the user's apparent intent.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding

The skill describes file-based behavior that reads and writes project documentation, but it does not declare any explicit tool scope or allowed-tools boundary. In an agent environment, missing capability scoping increases the chance of overbroad filesystem access or unintended writes, especially because the skill is designed to persist context across sessions.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill enables allow_implicit_invocation without any visible constrained trigger conditions, which can cause the agent to invoke a project-governance skill automatically in situations the user did not clearly request. Because this skill manages persistent project context and handoff notes, unintended invocation can propagate, alter, or expose sensitive project memory across tasks, increasing the chance of prompt-injection persistence or unauthorized context modification.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.