Model Context Protocol (MCP)

Give your AI agents direct access to AGM calendar data, board profiles, governance intelligence, and vote outcomes. Connect Claude, GPT, Gemini, and other AI assistants to AGM Monitor in seconds.

PRO

Paid Feature

MCP Tools

5,000+

Companies

Indices

REST API

Traditional programmatic access for dashboards, pipelines, and custom integrations.

Up to 25,000 requests / day

Standard REST endpoints
JSON responses
Build custom dashboards
Integrate into data pipelines
9 governance & calendar endpoints

View API Docs

MCP Server

AI-native access. Let your agents query AGM data, analyze governance, and monitor shareholder meetings autonomously.

Shares API rate limits · 25,000 req/day

Native AI agent integration
13 purpose-built tools
Works with Claude, GPT, Gemini & more
Structured data for LLM reasoning
Combine with other MCP servers
Agents ask the right questions for you

Contact Sales

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external data sources and tools. Instead of copying and pasting data into your AI chat, MCP lets your agent access AGM Monitor directly — querying meeting dates, analyzing board compositions, and reviewing vote outcomes on its own.

Connect

Add the AGM Monitor MCP server to your AI client using your API key.

Ask

Ask your AI agent questions like “When is Apple’s next AGM?” or “Show me Tesla’s board composition.”

Analyze

Your agent fetches live data and provides analysis, comparisons, and recommendations.

Quick Setup

Add the AGM Monitor MCP server to your AI client's configuration. Requires a paid API key.

// Add to ~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "agm-monitor": {
      "command": "npx",
      "args": ["-y", "@agmmonitor/mcp-server"],
      "env": {
        "AGM_API_KEY": "agm_live_your_api_key"
      }
    }
  }
}
    

// Add to .cursor/mcp.json in your project
{
  "mcpServers": {
    "agm-monitor": {
      "command": "npx",
      "args": ["-y", "@agmmonitor/mcp-server"],
      "env": {
        "AGM_API_KEY": "agm_live_your_api_key"
      }
    }
  }
}
    

# Run the MCP server directly via npx
AGM_API_KEY="agm_live_your_api_key" npx -y @agmmonitor/mcp-server

# Or install globally
npm install -g @agmmonitor/mcp-server
AGM_API_KEY="agm_live_your_api_key" agm-mcp-server

# The server communicates over stdio and is compatible with
# any MCP client: Claude Code, Windsurf, Zed, and more.
    

Compatible AI Clients

AGM Monitor's MCP server works with any client that supports the Model Context Protocol.

Claude Desktop

Anthropic

Claude Code

Anthropic CLI

Cursor

AI Code Editor

Windsurf

Codeium

Zed

AI Editor

Continue

Open Source

Cline

VS Code

Custom Agents

Any MCP client

Tool Reference

The MCP server exposes the following tools that your AI agent can call. All tools require a paid API key.

Resources (Read-Only Data)

RESOURCE agm://indices

Returns a list of all 20 supported indices with company counts and region information. This resource is automatically provided as context when your agent starts a session.

Example Output

Available indices: - S&P 500 (497 companies, US) - Russell 3000 (2,942 companies, US) - FTSE 100 (100 companies, UK) - FTSE 350 (350 companies, UK) - STOXX Europe 600 (600 companies, EU) - Nikkei 225 (225 companies, JP) ...

Tools (Agent-Callable Functions)

TOOL get_agm_calendar

Retrieve AGM calendar data for an entire index or a specific company. Supports filtering by year, sector, confidence level, and date range.

Parameter	Type	Description
index	string	Index ID: `sp500`, `russell3000`, `ftse100`, etc.
ticker	string?	Optional. Filter to a single company ticker.
year	number?	Filter by meeting year (e.g. 2026)
sector	string?	Filter by GICS sector
confidence	string?	Filter: `confirmed`, `scheduled`, or `predicted`

Example Agent Interaction

User: When is Apple's next annual meeting? Agent: Let me check the AGM calendar for Apple. [calls get_agm_calendar(index="sp500", ticker="AAPL")] Apple's 2026 Annual Meeting of Shareholders is scheduled for February 24, 2026. This is a confirmed date sourced from their 8-K filing with the SEC.

TOOL get_board_profile

Get the full board of directors for a company: names, roles, tenure, committee membership, independence status, and meeting attendance rates.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `AAPL`)

TOOL get_governance_timeline

Retrieve governance timeline events: record dates, 14a-8 shareholder proposal deadlines, director nomination windows, and earnings dates.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `MSFT`)

TOOL get_governance_practices

Get corporate governance structure: voting standards, proxy access terms, board classification, special meeting rights, and dual-class share status.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `GOOGL`)

TOOL get_vote_results

Proxy vote results for each ballot item: proposal descriptions, management recommendations, vote counts, percentages, and pass/fail outcomes.

Parameter	Type	Description
ticker	string	Company ticker
year	number?	Meeting year (default: most recent)

TOOL get_institutional_holders

Top 50 institutional holders with shares held, ownership percentage, quarter-over-quarter changes, voting authority breakdown, and 13D/13G status.

Parameter	Type	Description
ticker	string	Company ticker

TOOL get_proxy_voting

How top institutional investors actually voted on each ballot item, sourced from SEC Form N-PX filings. Includes management alignment analysis.

Parameter	Type	Description
ticker	string	Company ticker

TOOL get_compensation

Executive and director compensation: summary compensation tables, say-on-pay history, compensation committee details, stock ownership guidelines, pay-for-performance analysis, and CEO pay ratio.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `AAPL`)
section	string?	Optional. Filter: `summary`, `directors`, `say_on_pay`, `committee`, `pay_performance`

Example Agent Interaction

User: How much did Tim Cook make last year? Agent: Let me pull Apple's compensation data. [calls get_compensation(ticker="AAPL", section="summary")] Tim Cook's total compensation for FY2024 was $66.07 million: - Base salary: $3.0M - Stock awards: $50.0M - Performance incentive: $12.0M - Other compensation: $1.1M His say-on-pay vote received 90.2% support in 2025, up significantly from 64.7% the prior year.

TOOL get_news

Recent company news: headlines, summaries, sources, publication dates, and categories. Covers earnings, governance events, M&A, leadership changes, and more.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `AAPL`)
limit	number?	Max articles to return (default: 20)

TOOL get_activism

Activist campaigns targeting a company: activist identity, campaign timeline, demands, outcomes, stake percentages, related SEC filings, and media coverage.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `DIS`)
status	string?	Filter: `active`, `settled`, or `all` (default)

TOOL get_peer_benchmarking

Peer benchmarking: company's self-selected peer group with comparative compensation analysis, CEO pay ratios, say-on-pay support, and governance comparisons sourced from the DEF 14A.

Parameter	Type	Description
ticker	string	Company ticker (e.g. `AAPL`)

TOOL search_companies

Search for companies by name, ticker, or sector across all supported indices. Useful for finding the right ticker before querying specific data.

Parameter	Type	Description
query	string	Search term (company name, ticker, or sector)
index	string?	Optional. Limit search to a specific index.

TOOL compare_governance

Compare governance practices across multiple companies side by side. Returns a structured comparison of board composition, voting standards, proxy access, and shareholder rights.

Parameter	Type	Description
tickers	string[]	Array of 2-5 company tickers to compare

Example Agent Interaction

User: Compare Apple and Microsoft's governance structures. Agent: I'll compare governance practices for both companies. [calls compare_governance(tickers=["AAPL", "MSFT"])] Here's a side-by-side comparison: Apple Microsoft Board Size: 8 12 Independent Chair: Yes Yes Classified Board: No No Proxy Access: 3%/3yr/20% 3%/3yr/20% Voting Standard: Majority Majority Special Meeting: 10% threshold Not available Dual-Class Shares: No No Both companies follow strong governance practices. The key difference is Apple allows shareholders holding 10% of shares to call a special meeting, while Microsoft does not offer this right.

What Can Your Agent Do?

Here are some of the questions your AI agent can answer with AGM Monitor's MCP server.

Shareholder Engagement

“Which S&P 500 companies have AGMs in the next 30 days?”

“What’s the deadline to submit a shareholder proposal at Tesla?”

Governance Research

“Compare the board composition of FAANG companies.”

“Which tech companies have dual-class share structures?”

Proxy Analysis

“How did BlackRock vote on the executive compensation proposal at Apple?”

“Which proposals failed at Amazon’s last AGM?”

Authentication & Rate Limits

The MCP server authenticates using your API key, set via the AGM_API_KEY environment variable. MCP requests share the same rate limits as REST API calls.

# Set your API key as an environment variable AGM_API_KEY=agm_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxx

Limit	Value	Notes
Daily requests	25,000	Shared across API and MCP
Concurrent sessions	5	Max simultaneous MCP connections
Tools per session	All 13	All tools available in every session

Error Handling

When errors occur, the MCP server returns structured error messages that your AI agent can interpret and explain to you.

Error	Code	Description
UNAUTHORIZED	401	Missing or invalid API key in AGM_API_KEY
FORBIDDEN	403	API key does not have access to this tool
NOT_FOUND	404	Ticker or index not found
RATE_LIMITED	429	Daily request quota exceeded
SERVER_ERROR	500	Internal server error — retry or contact support