ClawHub
Back to skill

Security audit

Skill Test Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small documentation helper that reads target skill scripts and generates testing documentation without hidden network, credential, or persistence behavior.

Before installing, expect the agent to read scripts in the target skill and potentially create or edit TESTING.md or a Testing section in the target SKILL.md. Review generated documentation before committing it, especially because the helper emits templates rather than fully verified tests.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
84% confidence
Finding

The code substantially aligns with the general idea of generating test documentation from scripts, including normal, edge, and error cases plus command examples. However, the declared description claims it should be used specifically when a skill has scripts but lacks testing guidance, and says it creates or updates testing documentation. This implementation does neither: it never inspects TESTING.md or SKILL.md, and it does not write or update any documentation files—only prints generated markdown. Additionally, while it mentions analyzing scripts broadly, actual analysis is limited to Python scripts; shell scripts get fixed generic cases. These are material differences in capability, so this is a mismatch.

Content

No source excerpt is available for this finding.

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

md
- Test execution instructions
- Environment setup requirements

**Option B: Update SKILL.md**
Add a "Testing" section to the existing SKILL.md with:
- Overview of test approach
- Test case summaries

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.