Security audit

story writer

Security checks for vulnerabilities and agentic risk

Overview

This is a local fiction-writing workflow that creates project files, with no evidence of hidden network access, credential use, purchases, crypto activity, or destructive behavior.

Install only if you want a file-based writing workflow. Run it in a dedicated story folder, keep backups of drafts, and avoid granting purchase, crypto, credential, or network authority because the inspected artifacts do not need those capabilities.

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 (2)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The skill description is very broad ('Use when creating, planning, or writing fiction') and could cause the agent to invoke this multi-step pipeline for many ordinary writing requests where the user did not ask for project scaffolding, file generation, or persistent workflow management. In context, this increases the chance of unintended activation and unnecessary downstream actions such as creating project artifacts or steering the user into a heavyweight process.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill repeatedly describes saving outputs to persistent files across stages, but does not warn the user that invoking the workflow may create or overwrite files in a project folder. In this context, the risk is unintended filesystem modification, especially because the skill is sequential and encourages multi-stage artifact generation across many paths.

Static analysis

No suspicious patterns detected.