Peec AI MCP Server

Community-built MCP server for the Peec AI API — AI Search Analytics for brand visibility, sentiment, and citations across ChatGPT, Perplexity, and other AI models.

Note: This is an unofficial community project, not affiliated with or endorsed by Peec AI. It requires a Peec AI account and API key.

API Beta: The Peec AI API is currently in beta — endpoints, payloads, and responses may change. Access is limited to Enterprise customers.

What it does

Peec AI tracks how brands appear in AI-generated answers. This MCP server gives any MCP-compatible client direct access to that data — 38 tools covering projects, brands, prompts, chats, query analysis, analytics reports, source content, brand profiles, and full CRUD operations.

Key capabilities:

• Query brand visibility, sentiment, and position across AI models
• Analyze which domains and URLs get cited in AI responses
• Inspect individual chat interactions with full source attribution
• Slice data by model, prompt, category tag, topic, country, or date
• Create, update, and delete brands, prompts, tags, and topics (opt-in)
• Review and act on AI-generated prompt and topic suggestions

Quick Start

1. Get your API key

Sign up at app.peec.ai and create an API key under Settings > API Keys.

2. Configure

claude mcp add --transport stdio peecai -- npx -y mcp-server-peecai

export PEECAI_API_KEY="your-api-key"
export PEECAI_PROJECT_ID="your-project-id"  # optional default

{
  "mcpServers": {
    "peecai": {
      "command": "npx",
      "args": ["-y", "mcp-server-peecai"],
      "env": {
        "PEECAI_API_KEY": "your-api-key",
        "PEECAI_PROJECT_ID": "your-project-id"
      }
    }
  }
}

{
  "servers": {
    "peecai": {
      "command": "npx",
      "args": ["-y", "mcp-server-peecai"],
      "env": {
        "PEECAI_API_KEY": "your-api-key",
        "PEECAI_PROJECT_ID": "your-project-id"
      }
    }
  }
}

3. Verify

Confirm the peecai server is connected — in Claude Code run /mcp, in VS Code/Cursor check the MCP server status in the output panel.

Tools

Data Retrieval (19 tools)

list_projects — List all projects for the company.

• Returns: project IDs, names, statuses (CUSTOMER = active, PITCH = demo)
• Parameters: limit, offset

list_brands — List tracked brands with their associated domains.

• Parameters: project_id, limit, offset

list_prompts — List monitored search prompts.

• Returns: prompt messages, tags, topics, user location, search volume
• Parameters: project_id, topic_id, tag_id, limit, offset

list_tags — List category tags for a project.

• Parameters: project_id, limit, offset

list_topics — List topic groupings for a project.

• Parameters: project_id, limit, offset

list_models — Deprecated; prefer list_model_channels. List tracked AI models (ChatGPT, Perplexity, etc.).

• Returns: model IDs and active status
• Parameters: project_id, limit, offset

list_model_channels — List model channels (stable IDs grouping one or more models, e.g. openai-0, perplexity-0).

• Returns: channel ID, description, currently active model, active status
• Parameters: project_id, limit, offset

list_chats — List AI chat interactions with optional date and dimension filtering.

• Returns: chat IDs, prompt/model/model_channel refs, dates
• Parameters: project_id, start_date, end_date, brand_id, prompt_id, model_id, model_channel_id, limit, offset

get_chat_content — Get full content of a specific chat.

• Returns: sources (URLs, domains, citation counts), brands mentioned, messages, queries, products
• Parameters: chat_id, project_id

list_prompt_suggestions — List AI-generated prompt suggestions.

• Parameters: project_id, topic_id, limit, offset

list_topic_suggestions — List AI-generated topic suggestions.

• Parameters: project_id, limit, offset

list_brand_suggestions — List AI-generated brand suggestions.

• Returns: suggestion ID, name, domains, chat_count
• Parameters: project_id, limit, offset

get_project_profile — Read the project's brand profile (occupation, industry, brand identity, target markets, audience distribution).

• Returns: profile object or null if not yet profiled
• Parameters: project_id

Analytics Reports

All report tools support dimensions for multi-level breakdowns: prompt_id, model_id, model_channel_id, tag_id, topic_id, date, country_code, chat_id. Date filtering via start_date / end_date (YYYY-MM-DD). Server-side filtering via filters parameter (field, operator: "in" | "not_in", values).

get_brands_report — Brand analytics per brand.

get_domains_report — Domain-level analytics.

get_urls_report — URL-level analytics.

get_url_content — Get the scraped markdown content of a source URL discovered via get_urls_report.

• Returns: content (markdown, null while scraping pending), title, domain, channel_title, classification, url_classification, content_length, truncated, content_updated_at
• Parameters: url, project_id, max_length (1–20,000,000, default 100,000)
• If stored content exceeds max_length, truncated=true — re-request with a larger max_length to get more.

Query Analysis

search_queries — Get search queries AI models generated when answering prompts.

• Parameters: project_id, start_date, end_date, filters, limit, offset

shopping_queries — Get shopping/product queries AI models generated.

• Parameters: project_id, start_date, end_date, filters, limit, offset

Write Operations (19 tools, opt-in)

Write tools are disabled by default for safety. Enable them by setting PEECAI_ALLOW_WRITES=true.

When disabled, these tools are completely invisible — they don't appear in tools/list and cannot be called by any client.

set_project_profile — Replace the project's brand profile (occupation, industry, brand identity, target markets, audience distribution). Full overwrite; audienceDistribution percentages must sum to 100. Triggers a background refresh of prompt suggestions.

Delete operations are soft-deletes and irreversible through the API. Delete tools carry destructiveHint: true in their MCP annotations, causing clients like Claude Code to require explicit user approval before execution.

Tool Annotations

Resources

MCP resources provide reference data that clients can fetch without a tool call.

Resource templates support listing — clients can enumerate available resources across all projects.

Prompt Templates

Guided analytical workflows available as MCP prompts. All prompts support project_id autocompletion.

Example Prompts

"List my Peec AI projects"
"Show brand visibility for the last 30 days"
"Which domains get cited most in AI search results?"
"Compare brand sentiment across ChatGPT and Perplexity"
"Show me the full chat content for chat ID abc-123"
"Get URL report broken down by AI model and country"
"What search queries do AI models use when answering my prompts?"
"Create a brand 'My Brand' with domain mybrand.com"
"Add a new prompt: 'best CRM software 2025' for country DE"

Environment Variables

MCP Protocol Features

This server implements the MCP 2025-11-25 specification:

• Structured content — list tools return structuredContent alongside text for type-safe client parsing
• Tool annotations — readOnlyHint, destructiveHint, idempotentHint on every tool
• Progress notifications — report tools send progress updates when the client provides a progressToken
• Structured logging — API errors are sent as MCP log notifications with endpoint, status, and message context
• Prompt completions — project_id argument supports autocompletion via completable()
• Resource templates — with list callbacks for enumerating resources across projects
• Cancellation support — all tools forward the MCP AbortSignal to API calls

API Drift Detection

The Peec AI API is in beta and may change. A drift detection script compares the live OpenAPI spec against a committed snapshot:

npm run check:api-drift

• No drift: exit code 0, snapshot is current
• Drift detected: exit code 1, shows a diff of changes

No API key is required — the OpenAPI spec is publicly accessible.

Development

Prerequisites

• Node.js >= 22
• npm

Commands

npm install              # Install dependencies
npm run build            # Compile TypeScript to dist/
npm run dev              # Watch mode — recompile on changes
npm test                 # Run unit tests (372 tests)
npm run test:watch       # Run tests in watch mode
npm run test:integration # Run integration tests (requires PEECAI_API_KEY)
npm run check:api-drift  # Check for API spec changes

Integration Tests

Integration tests hit the live Peec AI API and are skipped by default in npm test.

# Read-only smoke test (all 19 read tools + prompts + resources)
PEECAI_API_KEY=xxx npm run test:integration

# Full CRUD round-trip (requires a test project + write access)
PEECAI_ALLOW_WRITES=true PEECAI_TEST_PROJECT_ID=or_xxx npm run test:integration

Project Structure

src/
├── index.ts              # Server entry point, tool/resource/prompt registration
├── api-client.ts         # HTTP client for Peec AI Customer API
├── types.ts              # TypeScript interfaces for API responses
├── schemas.ts            # Zod output schemas for structured content
├── util.ts               # Validation, date handling, MCP response helpers
├── prompts.ts            # MCP prompt templates (guided workflows)
└── tools/                # One file per MCP tool (or tool group)
    ├── projects.ts       # list_projects
    ├── brands.ts         # list_brands
    ├── prompts.ts        # list_prompts
    ├── tags.ts           # list_tags
    ├── topics.ts         # list_topics
    ├── models.ts         # list_models
    ├── model-channels.ts # list_model_channels
    ├── chats.ts          # list_chats
    ├── chat-content.ts   # get_chat_content
    ├── prompt-suggestions.ts  # list_prompt_suggestions
    ├── topic-suggestions.ts   # list_topic_suggestions
    ├── brand-suggestions.ts   # list_brand_suggestions
    ├── project-profile.ts # get_project_profile, set_project_profile
    ├── report-brands.ts  # get_brands_report
    ├── report-domains.ts # get_domains_report
    ├── report-urls.ts    # get_urls_report
    ├── url-content.ts    # get_url_content
    ├── queries-search.ts # search_queries
    ├── queries-shopping.ts    # shopping_queries
    ├── write-brands.ts   # create/update/delete brand
    ├── write-prompts.ts  # create/update/delete prompt
    ├── write-tags.ts     # create/update/delete tag
    ├── write-topics.ts   # create/update/delete topic
    └── suggestion-actions.ts  # accept/reject suggestions

License

MIT

Built by Tobias Hein at artaxo — a digital marketing agency specializing in AI Search Optimization.