> For the complete documentation index, see [llms.txt](https://docs.uptiq.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uptiq.ai/agent-builder/build/mcp-tools.md).

# MCP Tools

**MCP Tools** is the catalog of Model Context Protocol servers the platform makes available to your agent. Each entry is a server that exposes a set of related tools — pulling market data, scraping the web, retrieving from a knowledge base, querying a financial API. Connecting an MCP server adds its tool set to the agent's reach; the agent then chooses among those tools at runtime when a user's request calls for them.

Manage Agent → MCP Tools catalog

Reach this page from **Manage Agent → Integration Tools & Connectors → MCP Tools** in the left navigation.

### The catalog structure <a href="#the-catalog-structure" id="the-catalog-structure"></a>

The right pane organizes MCP servers into **categories**. The categories group servers by what they do, which is how you typically search: you're looking for a market-data source, or a scraping tool, or a search API, more often than you're looking for a specific server by name.

<table><thead><tr><th width="259.60003662109375">Category</th><th>What the servers in it do</th></tr></thead><tbody><tr><td>Analytics / Warehouses / BI</td><td>Market data, SEC filings, stock data, financial APIs</td></tr><tr><td>Web, Browsing &#x26; Scraping</td><td>Content extraction, crawling, programmatic web access</td></tr><tr><td>General Search &#x26; Knowledge Retrieval</td><td>Search APIs, knowledge-base retrieval</td></tr></tbody></table>

The catalog likely extends beyond what was visible in capture; use the right-pane search (`Search MCP tools…`) to find a server by name if you know which one you want.

#### Example entries <a href="#example-entries" id="example-entries"></a>

A few representative servers from the catalog:

* **CoinAPI Market Data API** — real-time and historical normalized trades, order books, OHLCV, tick-level history, multi-exchange coverage.
* **Finfeed SEC API** — clean, structured SEC filing data through a developer-friendly REST API.
* **Finfeed Stock API** — T+1 raw stock market data, IEX, Euronext, and 20+ stock exchanges.
* **Alpha Vantage API** — real-time and historical financial market data including stocks, forex, crypto, and economic indicators.
* **Firecrawl** — LLM-friendly web scraping for content extraction, crawling, and search.
* **Tavily** — search and data extraction designed for AI assistants, exposed via MCP.
* **AWS Knowledge Base Retrieval** — retrieves from AWS Knowledge Base using the Bedrock Agent Runtime, enabling Retrieval-Augmented Generation patterns.

> **Note — Catalog not fully enumerated** Categories below the visible scroll were not enumerated during capture. The catalog likely contains additional categories and servers beyond what's listed here. Use the right-pane search to find specific servers.

### Connecting an MCP server <a href="#connecting-an-mcp-server" id="connecting-an-mcp-server"></a>

The connect interaction follows the same pattern as Apps & Services: hover the row to reveal the action, click to begin the connection flow, supply whatever the server requires (API keys, OAuth, endpoint URLs).

> **Note — Depth gap** The MCP connect-flow detail (credentials format, configuration options, per-tool authorization within a server) was not exercised during capture. Coverage is queued for a future pass.

### Where connected MCPs show up <a href="#where-connected-mcps-show-up" id="where-connected-mcps-show-up"></a>

After connecting, the MCP server appears as a node on the Layout canvas — one of the disc-shaped icons in the Skills band labeled with the server's name (the example agent shows `accountancy`, `credit-score`, `general`, `analytics`, `banking`). The server's tools become available to the agent through the Skills & Capabilities page.

### Pre-existing connections <a href="#pre-existing-connections" id="pre-existing-connections"></a>

In an example agent, several MCPs may already be attached when you arrive. These were added either by your team or by the platform during agent provisioning. Inspect them by hovering their connector nodes on the Layout canvas; manage them by opening this page or Skills & Capabilities.

### MCP Tools compared to Apps & Services <a href="#mcp-tools-compared-to-apps-and-services" id="mcp-tools-compared-to-apps-and-services"></a>

The catalogs overlap. The rule of thumb:

* If the action involves a *recognized SaaS product* with a logo and a stable customer-facing API (Stripe, Resend, Twilio), start in Apps & Services.
* If the action involves a *data API or a generic tool* (market data, web scraping, knowledge-base retrieval), start in MCP Tools.
* If both catalogs contain an entry for the service you want, prefer the one that fits your authentication situation — some teams already have OAuth set up for an Apps & Services entry, while others have a service API key handy that an MCP entry can use directly.

### Why this matters <a href="#why-this-matters" id="why-this-matters"></a>

MCPs are how the agent reaches the long tail of data and capability the platform doesn't ship as a first-party connector. A general-purpose reasoning model becomes a *useful* agent for a specific domain when it can read market data your team trusts, search the web with a paid service that filters out junk, and retrieve from your organization's knowledge base — and MCPs are how each of those reaches the agent's hands. Treat MCP Tools as the inventory of the agent's worldly capabilities; every server you connect here is one more domain the agent can act within.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.uptiq.ai/agent-builder/build/mcp-tools.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.