ClawHub
Back to skill

Security audit

Skill Creator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed workflow helper for creating and evaluating other skills, with some operational risk from file writes, subagents, and a local review viewer but no evidence of hidden or malicious behavior.

Install this only if you want the agent to help build and test skills. Expect it to create workspace files, spawn evaluation runs, and possibly open or generate a local review viewer; review generated skills before installing or sharing them, and stop any viewer process after use if it was launched.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill explicitly advises making descriptions "a little bit pushy" and to trigger on broad, common contexts even when the user does not explicitly ask for the capability. That can cause over-triggering, pulling this skill into unrelated tasks and increasing the chance of inappropriate file operations, background execution, or other side effects from a powerful skill being invoked when it should not be.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
88% confidence
Finding

The instruction to launch the viewer with nohup starts a detached background process that persists beyond the immediate interaction. Persistent processes can consume resources, leave unintended services running, and expose generated review artifacts longer than necessary, especially if cleanup is missed or the environment is shared.

Content

Scanner excerpt · SKILL.md (reported line 238)May include surrounding context.

  1. Launch the viewer with both qualitative outputs and quantitative data:
    bash
    nohup python <skill-creator-path>/eval-viewer/generate_review.py \
  <workspace>/iteration-N \
  --skill-name "my-skill" \
  --benchmark <workspace>/iteration-N/benchmark.json \

Static analysis

No suspicious patterns detected.