# sharozdawa/ai-visibility [Health: Active]

**Category:** 🎯 Marketing  
**Repository:** https://github.com/sharozdawa/ai-visibility  
**GitHub Stars:** 11  
**npm Downloads (last month):** 186  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/sharozdawa-ai-visibility

## Description
Track brand visibility across ChatGPT, Perplexity, Claude, and Gemini. Visibility scores, sentiment analysis, competitor detection, and trend charts.

## Tools
Capabilities this server exposes over MCP:

- **check_brand_visibility** — Check a brand's visibility across AI platforms (ChatGPT, Perplexity, Claude, Gemini). Simulates realistic queries and analyzes mention rates, positions, sentiment, and competitor landscape.
- **check_single_query** — Check if a brand is mentioned for a specific query on a specific AI platform. Returns mention status, position, context snippet, sentiment, and competitor mentions.
- **get_visibility_score** — Calculate an overall AI visibility score (0-100) for a brand across all four AI platforms. Includes per-platform breakdowns and improvement recommendations.
- **compare_brands** — Compare the AI visibility of multiple brands side by side. Shows per-platform scores, overall rankings, and relative strengths.
- **get_recommendations** — Get actionable recommendations to improve a brand's AI visibility. Prioritized suggestions based on current score and industry best practices.
- **list_platforms** — List all supported AI platforms with details about how each sources and presents brand information.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "ai-visibility": {
    "command": "npx",
    "args": ["-y","ai-visibility-mcp"]
  }
}
```

## Documentation

## What sharozdawa/ai-visibility MCP server does

The sharozdawa/ai-visibility MCP server provides tools for evaluating brand visibility in four AI platforms: ChatGPT, Perplexity, Claude, and Gemini. It focuses on whether a brand is mentioned, where it appears, how it is described, and which competing brands appear in the same results.

The available analysis covers both broad brand checks and individual query checks. A broad check returns platform-level results, mention rates, positions, sentiment, and competitor information. A single-query check narrows the result to one query and one platform, returning the mention status, position, context snippet, and sentiment.

## How it works

The server calculates an overall visibility score from 0 to 100 using three weighted inputs:

- Mention rate contributes 40 percent.
- Position quality contributes 30 percent.
- Sentiment contributes 30 percent.

Scores are grouped into five tiers: Excellent, Good, Moderate, Low, and Very Low. The scoring model treats earlier mentions as better positions, with first place ranked above later positions. Sentiment is categorized as positive, neutral, or negative.

The sharozdawa/ai-visibility MCP server also provides side-by-side comparison for two to ten brands. Its platform listing describes how each supported AI system sources and presents brand information. Results use seeded randomness, so the same brand can produce reproducible results rather than changing unpredictably between identical runs.

## Setup and configuration

The quickest setup uses the published npm package through npx:

```bash
npx ai-visibility-mcp
```

For Claude Desktop, add an MCP entry that runs `npx` with `-y ai-visibility-mcp` as its arguments. Cursor uses the same command structure in `.cursor/mcp.json`.

Self-hosting is also supported. Clone the repository, enter `mcp-server`, install dependencies, build the project, and run `node dist/index.js`. An MCP client can then be configured to launch that local JavaScript file instead of using npx.

The repository additionally includes a Next.js dashboard. Running that dashboard requires a `.env` file with `DATABASE_URL`, Prisma database setup, and the project’s development command; this configuration is separate from the documented MCP server launch.

## Tools and capabilities

The server exposes six tools:

- `check_brand_visibility` reviews a brand across supported platforms and returns detailed visibility, position, sentiment, mention-rate, and competitor data.
- `check_single_query` checks one brand, query, and platform for mention status, position, context, and sentiment.
- `get_visibility_score` returns the 0–100 score, tier, platform breakdowns, and improvement recommendations.
- `compare_brands` ranks two to ten brands and provides per-platform scores.
- `get_recommendations` produces prioritized actions based on the current visibility score.
- `list_platforms` describes the supported platforms and their information sources and presentation models.

## Limitations and notes

The documented platform set is limited to ChatGPT, Perplexity, Claude, and Gemini. Brand comparisons accept between two and ten brands. Because the project uses seeded randomness for consistent results, outputs should be understood as reproducible tracker results rather than an indication that every response from an AI platform has been continuously monitored.

_Full upstream README: https://allmcps.com/mcp/sharozdawa-ai-visibility/readme_