seo-dataforseo

• Outcome: Pass
• Latest audit: Aug 12, 2026 · 2:39 PM UTC
• Version: 1.0.0

Overview

This skill is a disclosed DataForSEO integration for live SEO research, with paid API and scraping-related capabilities that users should invoke deliberately.

Install this only if you intend to use DataForSEO and are comfortable sending keywords, domains, URLs, and SEO queries to that service. Review your DataForSEO budget settings, confirm expensive scraper or backlink operations, and specify location/language when US English is not intended.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The trigger phrases are broad enough that ordinary user requests such as 'keyword volume', 'backlink data', or 'real search data' could invoke this skill unexpectedly. Because the skill performs paid external lookups and can reach many MCP tools, unintended invocation can cause unnecessary data access and API spend even without explicit user intent.

Natural-Language Policy Violations

Low

• Confidence: 80% confidence
• Finding: Defaulting to US/English without user opt-in can cause queries to be sent to an unintended geography or language context. This is primarily a consent and correctness issue, and in some cases may expose user intent to a third-party service under assumptions the user did not choose.

Natural-Language Policy Violations

Low

• Confidence: 82% confidence
• Finding: The SERP command hard-codes US/English defaults, which may silently route searches through a jurisdiction or language context the user did not request. While not directly enabling code execution or privilege escalation, it can produce privacy, consent, and data-quality issues.

Natural-Language Policy Violations

Low

• Confidence: 82% confidence
• Finding: The image SERP command uses implicit US/English defaults, which can cause unexpected third-party queries and inaccurate localization. This is a low-severity issue but becomes more relevant because the skill is user-invocable and external-service backed.

Natural-Language Policy Violations

Low

• Confidence: 82% confidence
• Finding: Keyword idea generation defaults to US/English without explicit user choice, potentially misrepresenting market data and sending user requests under the wrong locale. The risk is low but real because results may influence decisions and incur paid API calls.

Unrestricted Tool Access

Medium

• Category: Excessive Agency

Content

Run on-page analysis including Lighthouse audit and content parsing.

**MCP tools:** `on_page_instant_pages`, `on_page_content_parsing`, `on_page_lighthouse`

**Usage:**
- `on_page_instant_pages`:Quick page analysis (status codes, meta tags, content size, page timing, broken links, on-page checks)

• Confidence: 77% confidence
• Finding: The on-page analysis command can fetch and analyze arbitrary user-supplied URLs, including Lighthouse and content parsing operations that may perform network requests and deeper crawling. In an agent context, that can be abused for unintended access to internal or sensitive endpoints, SSRF-like behavior, or costly scans if runtime URL restrictions are not enforced.

Unrestricted Tool Access

Medium

• Category: Excessive Agency

Content

Scrape what ChatGPT web search returns for a query. ChatGPT visibility check: see which sources ChatGPT cites for your target keywords. Check Google AI Overviews and AI Mode with GSC gen-AI reports when available.

**MCP tools:** `ai_optimization_chat_gpt_scraper`

**Parameters:** query, location_code (optional), language_code (optional). Use `ai_optimization_chat_gpt_scraper_locations` to look up available locations.

• Confidence: 84% confidence
• Finding: The ChatGPT scraper capability introduces a scraping/fetching surface against an external service and may process user-supplied queries that trigger automated retrieval of third-party content. In an agent environment, this raises risk of policy bypass, unreviewed external collection, or abusive/expensive automated scraping if not strongly constrained.

Static analysis

No suspicious patterns detected.