ClawHub
Back to skill

Security audit

OGT Docs

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation workflow guide with no bundled code, network behavior, credential use, or hidden execution.

Install only if you want agents to follow a docs-first project workflow. Review its convention that documentation overrides code, because it may lead agents to propose changing implementation to match docs; keep normal review before applying meaningful code or project-structure changes.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
1. **If it's not documented, it doesn't exist**
2. **If code contradicts docs, code is wrong**
3. **Never trust "done" status without verification**
4. **Move folders, don't copy files**
5. **Signal with dot-files, don't edit status fields**
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Static analysis

No suspicious patterns detected.