tutorial-viral-f100k

• Outcome: Review
• Latest audit: Sep 17, 2026 · 4:20 PM UTC
• Version: 1.0.0

Overview

The skill mostly supports creating faceless tutorial kits, but its instructions conflict about local video rendering and include install and privileged troubleshooting commands that need careful review.

Install only if you want this agent to help assemble tutorial-kit assets. Treat the local render, Homebrew, bootstrap, and sudo npm commands as manual review items rather than automatic steps, and avoid invoking /tutorial-render unless you intentionally want local environment changes and understand which external scripts will run.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 89% confidence
• Finding: The manifest description at L003 limits the skill to producing a kit and explicitly says it does not render or edit the video in this agent. However, the body of the skill describes a full Phase 4 render pipeline, final outputs including `BORRADOR_AUTO.mp4`, and opening the rendered file, which expands behavior beyond the manifest's stated scope.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The activation list includes broad phrases like 'hazme un tutorial de X' and 'video sin cara', which can match many ordinary requests outside this skill's intended narrow workflow. Although some exclusions are listed, the trigger set still lacks enough specificity to reliably distinguish this skill from other tutorial or video-generation skills.

Intent-Code Divergence

Medium

• Confidence: 92% confidence
• Finding: Lines L014-L017 explicitly say Phase 4 does not run here and that the render scripts are not included in this package. Later sections nevertheless describe `/tutorial-render`, direct execution of `render_tutorial.py`, bootstrapping `editor-video-formula100k`, outputting `BORRADOR_AUTO.mp4`, and troubleshooting local render errors, which contradicts that earlier documentation rather than merely omitting detail.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The 'Cuándo activar' section says to activate when the user asks for a tutorial or mentions faceless/screen recording terms, but these conditions are broad and overlap with common conversational intents. This ambiguity increases the risk of unintended invocation when a user merely discusses tutorial-style content rather than requesting this full pipeline.

Sudo/Root Execution

Medium

• Category: Privilege Escalation

Content

| `render_tutorial.py: voz no encontrada` | Verificar que el archivo exista en `$DEST/USER/` con el nombre exacto indicado |
| Keyword no aparece en transcript | Editar el MANIFEST con una palabra que sí esté en el transcript (ver `captions.json`) |
| `bootstrap.sh` reporta Homebrew faltante | Instalar Homebrew con el comando oficial y reintentar |
| `npm install` falla en Remotion | Correr `sudo chown -R $(whoami) ~/.npm` y reintentar |
| Captions desfasados de la voz | Drift de Whisper — raro; borrar `captions.json` y re-render fuerza re-transcripción |
| Pantalla dura menos que su segmento | El motor hace loop/hold del último frame; ajustar duración en MANIFEST si se ve raro |

• Confidence: 97% confidence
• Finding: The skill recommends running 'sudo chown -R $(whoami) ~/.npm' as a troubleshooting step. Recursive ownership changes under elevated privileges are risky because they normalize privileged shell execution, can mask deeper permission problems, and may damage local environment integrity or be adapted into more dangerous privileged filesystem modifications.

Static analysis

No suspicious patterns detected.