sign in
sign in
// // autogen integration

Market data tools for
AutoGen agents.

TickerDB exposes pre-computed financial data as MCP tools. AutoGen's MCP support lets your agents call market data tools directly — trend analysis, valuation checks, ticker event history, and more.

try for free read the docs
// setup

Connect in a few lines.

AutoGen supports MCP tool servers natively via SseMcpToolAdapter. Point it at TickerDB's remote MCP server, and your agents get access to all available market data tools.

python 
# Connect AutoGen to TickerDB's MCP server
from autogen_ext.tools.mcp import SseMcpToolAdapter, SseServerParams

server_params = SseServerParams(
    url="https://mcp.tickerdb.com/",
    headers={"Authorization": "Bearer tapi_your_api_key"},
)

adapter = SseMcpToolAdapter(server_params=server_params)
tools = await adapter.get_tools()

# Every tool is now available to your agents
// agent workflow

Multi-agent market analysis.

AutoGen's multi-agent conversations get richer with real market data. Give your agents TickerDB tools and they can pull summaries, use summary event mode for historical context, and monitor watchlists — passing context between agents automatically.

python 
# Give an AutoGen agent market data tools
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.tools.mcp import SseMcpToolAdapter, SseServerParams

server_params = SseServerParams(
    url="https://mcp.tickerdb.com/",
    headers={"Authorization": "Bearer tapi_your_api_key"},
)

adapter = SseMcpToolAdapter(server_params=server_params)
tools = await adapter.get_tools()

agent = AssistantAgent(
    name="market_analyst",
    model_client=model_client,
    tools=tools,
)

result = await agent.run(
    task="Get a summary of AAPL and check its historical oversold events"
)
// state changes

Track state changes effortlessly.

Most market data APIs return point-in-time snapshots. TickerDB tracks state transitions — your agent sees what changed, not just what is.

python 
# Ask the agent to check for state changes
result = await agent.run(
    task="Check my watchlist for state changes and summarize what moved"
)
json 
{
  "ticker": "AAPL",
  "changes": [
    {
      "field": "rsi_zone",
      "from": "neutral",
      "to": "oversold"
    },
    {
      "field": "trend",
      "from": "uptrend",
      "to": "downtrend"
    }
  ]
}
// what your agent sees

What your agent sees.

Every tool returns categorical facts — not raw OHLCV data. Your agent can branch on "oversold" without needing to know what RSI > 70 means.

json 
{
  "ticker": "NVDA",
  "trend": "strong_uptrend",
  "momentum": {
    "rsi_zone": "overbought",
    "macd_signal": "bullish"
  },
  "volatility": "high",
  "fundamentals": {
    "pe_zone": "above_historical_avg",
    "earnings_surprise": "positive"
  }
}
// available tools

What your agent can call.

Every tool returns categorical, pre-computed data. Your agent gets facts it can reason about immediately.

get_summary

Full categorical snapshot for a single asset — trend, momentum, volatility, volume, extremes, fundamentals, support/resistance.

get_watchlist

Live summary data for all tickers in your saved watchlist.

get_watchlist_changes

Field-level diffs for your watchlist since the last pipeline run.

add_to_watchlist

Add tickers to your persistent watchlist.

remove_from_watchlist

Remove tickers from your watchlist.

get_search

Multi-field filtering across all assets. Build complex queries with arbitrary filter combinations.

get_schema

All queryable fields with types, values, and descriptions. Always free.

get_account

Your plan tier, rate limits, and current API usage.

create_webhook

Register a webhook URL for watchlist change notifications.

list_webhooks

List your registered webhook URLs.

delete_webhook

Remove a registered webhook.

full tool reference →
// why tickerdb + autogen

Built for how agents actually work.

Categorical, not numerical

TickerDB returns "rsi_zone": "oversold" instead of raw RSI values. Your AutoGen agents reason on categories they already understand — no prompt engineering required.

Pre-computed

Our data is computed once daily after market close and cached. Your agents get instant responses with zero request-time computation.

Tiny context footprint

A TickerDB response uses a fraction of the tokens you'd need to pass raw OHLCV data. More room for reasoning, less spent on input.

Start building.

Try for free. No credit card required.

try for free read the docs