ClawHub
Back to skill

Security audit

The Storyteller's Workbench

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only literary fiction coaching skill with some broad routing language, but no hidden execution, data access, persistence, or unrelated authority.

Before installing, understand that this skill may activate for a fairly wide range of fiction-writing requests and gives direct, craft-focused feedback that may include dark or uncomfortable literary material. It does not appear to run code, access files, use credentials, or persist 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
  • 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
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger description is extremely broad and includes catch-all phrasing like 'anything involving storytelling craft,' which can cause this skill to activate for a wide range of loosely related writing requests. Over-broad activation increases the chance of misrouting user tasks, overriding more appropriate skills, and causing the system to apply this skill's strong behavioral instructions in contexts it was not intended for.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The invocation guidance does not clearly separate this skill from neighboring domains such as general writing assistance, brainstorming, editing, prose generation, or non-literary fiction support. Without firm boundaries, the router may select this skill for unrelated or marginal tasks, leading to policy drift, degraded handling, or inappropriate application of instructions like 'never to sanitise' and 'be willing to go dark.'

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The instruction to load this file when the user's request is "advanced" or when main techniques "aren't sufficient" does not define clear boundaries for activation. Because this is a markdown file, vague invocation guidance can cause the skill content to be pulled in too often or inconsistently, with no explicit scope or exclusion examples.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.