ClawHub
Back to skill

Security audit

Seo Audit

Security checks for vulnerabilities and agentic risk

Overview

This SEO-audit skill is a coherent guidance-only skill, with a small privacy caution around reading a specific local marketing context file.

Before installing, be aware that the skill may read a local product marketing context file if one exists and may use external SEO tools on URLs you provide. Avoid putting confidential strategy in those context files unless you are comfortable with the agent using it during SEO work.

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

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill description contains many broad trigger phrases such as 'help with SEO' and 'my SEO is bad,' which can cause the skill to activate for vague or overlapping requests that may be better handled by other skills. This increases the chance of incorrect tool use or unnecessary data access during an audit workflow, especially because the skill later instructs the agent to inspect local context files and potentially external URLs.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly instructs the agent to read a local file such as '.agents/product-marketing-context.md' before asking questions, without requiring user awareness or consent. That creates a project-data exposure risk because the agent may ingest sensitive local business, product, or strategy information unrelated to the user's immediate request and potentially surface it in its response.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The eval explicitly expects the skill to trigger on the casual phrase 'why am I not ranking,' which broadens activation beyond a tightly scoped SEO-audit request. Over-broad routing can cause the wrong skill to intercept ambiguous user intents, leading to irrelevant guidance, privacy over-collection (for example asking for Search Console access when unnecessary), or blocking a more appropriate specialized skill.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.