Back to skill

Security audit

dark-editorial-retouch

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: its findings are low concern and appear related to its stated use, but missing references leave some evidence incomplete.

The verdict is benign, not a guarantee of safety. Consider that the activation phrases may be broad, and that Skillspector could not resolve several references, including references to scripts and assets; the available evidence is therefore incomplete.

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
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)

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/github-skills-research.md (reported line 16)May include surrounding context.

md
visual y recien despues producir. **Aplicado:** la skill define "clave baja editorial"
  como tesis y deriva todas las perillas de ahi; el texto de la marca queda como unico
  acento grafico.
  https://github.com/anthropics/skills/blob/main/skills/canvas-design/SKILL.md
- `skills/brand-guidelines` — paleta y tipografia como contrato, con variantes por
  contexto. **Aplicado:** `assets/brand/` con 5 variantes de la insignia (knockout,
  lineas, color, monocromo oscuro, tamanos de avatar) y la regla de cuando usar cada una;

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/github-skills-research.md (reported line 21)May include surrounding context.

md
contexto. **Aplicado:** `assets/brand/` con 5 variantes de la insignia (knockout,
  lineas, color, monocromo oscuro, tamanos de avatar) y la regla de cuando usar cada una;
  la insignia a color nunca se recolorea (ver `references/logo-integration.md`).
  https://github.com/anthropics/skills/blob/main/skills/brand-guidelines/SKILL.md

## 2. `ComposioHQ/awesome-claude-skills` (77k ★)

The activation examples include broad phrases such as "mas profundidad", "brillos", and "cinematografico" without explicitly requiring the skill's photo-retouching context.

Low
Category
Trigger Abuse
Confidence
88% confidence
Finding

These phrases can occur in unrelated requests about writing, illustration, animation, or other visual work. Although the surrounding description identifies a specialized retouching purpose, this activation bullet presents the phrases as sufficient conditions without tying them to suitable photographs.

Content

No source excerpt is available for this finding.