ClawHub
Back to skill

Security audit

creator-rights-assistant

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation assistant for creator rights records, with privacy and consent limits clearly stated.

Before installing, understand that the records this skill helps create can contain sensitive rights, collaborator, licensing, and dispute context. Keep sidecar files under your control, use minimal exports unless there is a clear reason to share more, and get legal counsel for rights questions with legal weight.

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 (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Portability and capability assumptions

The sidecar is the state. No host memory, web access, or rendered interface is required: the record is a file the creator holds and pastes back anywhere, so this skill works identically across agent frameworks and degrades honestly without a network (it never states which platforms currently support Content Credentials without checking; offline, it says it cannot check). Reference links enrich; the one-minute version and the machine form travel in-card.

## The record knows where it stands
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## What this skill never does

Validates licenses or permissions. Assesses ownership or infringement. Drafts legal documents. Predicts platform actions or dispute outcomes. Certifies authorship or dates. Embeds identity into files without consent. Watches a catalog or nudges about it unprompted.

## Worked example, as a transcript
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.