A comprehensive market research platform with MCP servers for Google search, web scraping, YouTube transcripts, and X (Twitter) data. Provides tools for competitive analysis, market gap analysis, industry research, and sales forecasting.
Server has 9 tools with basic schemas and descriptions, but significant quality gaps. Descriptions are present but generic (avg ~100 chars, below 194-char baseline). No output schemas documented. Error handling is minimal, most tools return raw exception strings rather than actionable recovery guidance. Tool names follow verb_noun convention (good), but parameter naming is inconsistent (e.g., 'query' vs 'username' vs 'video_id'). No tool annotations (readOnlyHint, destructiveHint). Composition is reasonable, each tool has one responsibility, but chaining support is weak (no IDs returned to enable downstream calls).
Fetch profile details and public metrics (followers, following, tweet count, bio) for an X handle. Args: username (str): The X handle without '@' (e.g., 'Google', 'Microsoft') Returns: str: Detailed profile metrics and information for the user
Fetch recent tweets posted by a specific X (Twitter) handle (e.g. brand, competitor, or executive). Args: username (str): The X handle without '@' (e.g., 'OpenAI', 'elonmusk') max_results (int): Number of tweets to fetch (default: 10, max: 100) Returns: str: Analysis of recent tweets from the specified X user
Search Google for market research, trends, and general information
Get Google Trends data and summaries for market trend analysis
Scrape a website and return its content in clean markdown format. Args: url (str): The URL to scrape include_links (bool): Whether to include links in the markdown (default: True) include_images (bool): Whether to include images in the markdown (default: False) Returns: str: Website content in markdown format with analysis summary
Parameter descriptions missing or incomplete. 'query' param in google_search, search_google_shopping, search_google_news lacks guidance on format, length, or constraints. 'max_results' in X tools lacks min/max bounds.
No output schemas documented. Tools return free-text strings (e.g., 'Formatted analysis of recent X tweets') without specifying structure, fields, or format. LLMs cannot plan downstream calls or extract structured data.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 66 | <=2025-11-25 | v2 |
Search Google News for recent news articles and market updates
Search Google Shopping for product pricing and availability
Search recent tweets on X (formerly Twitter) for market research, trends, sentiment, or brand mentions. Args: query (str): The search term or topic on X max_results (int): Number of results to return (default: 10, max: 100) Returns: str: Formatted analysis of recent X tweets with metrics
Fetch and summarize a YouTube video's transcript for content analysis. Args: video_id (str): The YouTube video ID to get transcript from Returns: str: Summarized transcript content
Error handling returns raw exception strings (e.g., 'Error scraping website {url}: {str(e)}') without recovery guidance. X tools have better error handling with fallback logic, but most tools lack actionable error messages.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All tools are read-only, but this is not declared in the schema. Per current MCP spec (2026-07-28), tool annotations enable better agent planning and safety.
Inconsistent parameter naming across tools. 'query' (google tools), 'username' (X tools), 'video_id' (YouTube), 'url' (scraper). Consider standardizing on 'search_query', 'user_handle', 'resource_id'.