ClawHub
Back to skill

Security audit

content-review-handling-desk

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small text-only workflow for content handling guidance and review planning, with no code, credentials, persistence, or private-file access.

Before installing, consider whether the broad phrase 'routine content handling work' is acceptable for your agent's skill-selection behavior; otherwise this appears proportionate for content review guidance and planning.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description and activation guidance are broad enough to match ordinary 'content handling' or 'content work' requests without strong scoping constraints. In an agent system that auto-selects skills from natural language, this can cause unintended invocation, leading the agent to apply this workflow in contexts the user did not explicitly intend.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The Stage 1 trigger condition says to use the skill for 'routine content handling work' when the user asks to set handling guidance, but it does not define what requests qualify or what nearby intents should be excluded. This ambiguity increases the chance that loosely related requests are routed into the tool, producing handling decisions from incomplete or mismatched context.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The Stage 2 trigger condition is similarly underspecified and may match any request to 'prepare a content review plan' even when the necessary validated `handling_guidance` is absent or informal. This can cause the agent to fabricate or infer review constraints, which may misroute sensitive content or generate an unsafe review workflow.

Static analysis

No suspicious patterns detected.