ClawHub
Back to skill

Security audit

Skill Builder

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward guide for creating OpenClaw skills, with persistence clearly tied to the user's intent to make reusable skills.

Before installing, understand that this skill helps agents create persistent reusable skills. Review any generated SKILL.md before it is saved under a workspace or personal skills directory, especially if the proposed skill would use credentials, network access, shell commands, or private files.

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 (4)

Self-Modification

High
Category
Rogue Agent
Confidence
96% confidence
Finding

This skill explicitly instructs the agent to write a new SKILL.md, which is a self-modification capability because it changes the agent's reusable behavior surface. Self-modification is dangerous because it can persist mistakes, unsafe instructions, or adversarially influenced workflows beyond the current conversation.

Content

Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.

The Skill Creation Loop

text
Observe pattern → Define skill → Write SKILL.md → Test → Refine

Phase 1: Observe & Define

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 207)May include surrounding context.

md
- Skills are hints, not guarantees — always verify before acting on memory
- Code identifiers and paths stay in native form regardless of surrounding language
- Write skill entries in the user's language (中文/English)
- If user language is mixed, keep natural language entries consistent with surrounding content

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description uses broad trigger phrases like creating a skill or turning a workflow into a skill, which can match many normal user requests without requiring explicit confirmation. This can cause the agent to invoke a self-modifying capability too eagerly, increasing the chance of unintended skill generation or persistence of unsafe workflows.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill may activate when the agent merely 'notices' a workflow repeating 3+ times, even if the user never asked for automation or persistence. That ambiguity is risky because it lets the agent escalate from assisting with tasks to proposing or performing durable behavior changes based on inferred intent.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.