ClawHub
Back to skill

Security audit

Shopify Blog SEO Optimizer

Security checks for vulnerabilities and agentic risk

Overview

This skill has real Shopify article write capability, but it is disclosed, scoped to blog body/summary updates, and gated behind preview and explicit approval.

Install only for a store where you are comfortable granting Shopify content read/write access. Review the generated HTML report and exact approved fields before allowing --execute, and keep client secrets or automation tokens in the private env file, not chat or reports.

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (1)

Self-Modification

High
Category
Rogue Agent
Content
# Shopify Blog SEO Optimizer

This is an Admin write skill with preview, content audit, research, and post-write verification.

## What the merchant experiences
Confidence
87% confidence
Finding
This skill is explicitly capable of modifying Shopify article content after approval, so it has real write-side effects on a production storefront. Although the write capability is intended and constrained, any agent skill that can alter remote content increases risk if approval checks are bypassed, target selection is ambiguous, or generated content is unsafe or inaccurate.

Static analysis

Detected: suspicious.env_credential_access, suspicious.exposed_secret_literal

Environment variable access combined with network send.

Critical
Code
suspicious.env_credential_access
Location
scripts/lib/shopify-dev-dashboard-auth.mjs:34

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
references/onboarding-guide.md:45

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:118