Overview

MCP Server

Connect AI coding assistants and agents to the TextReveal API using the Model Context Protocol, with tools for every API route and built-in data context.

The TextReveal MCP server gives AI assistants and coding agents direct access to the TextReveal API. It exposes every API route as a callable tool and includes context to help the model understand your entities, universes, ESG/SDG indicators, and document data. The server runs at https://mcp.textreveal.com/mcp using streamable HTTP with OAuth.

TextReveal documentation is available in LLM-friendly formats. Append .md to any documentation page URL to retrieve the page as plain markdown. A structured index of the full site is available at /llms.txt, suitable for agents building context about the API before making calls.

Example queries

Once connected, you can ask your AI assistant questions such as:

"What are the latest ESG events for Apple over the last 30 days?"
"Which entities in my universes had the most environmental controversies this week?"
"Show me the ESG score trend for Tesla over the past year."
"Find all entities matching the keyword 'renewable energy' in my universes."
"What ESG categories apply to an event about a water pollution fine?"

The server understands the relationships between entities, events, documents, and indicators, so you can ask follow-up questions or chain requests naturally across multiple API routes.

Resources

The server exposes MCP resources that AI assistants can read for additional context. Documentation resources are the markdown versions of this site's dictionary pages, fetched at startup so the model has direct access to TextReveal concepts:

Resource URI	Dictionary page
`docs://entities`	Entity
`docs://events`	Event
`docs://universes`	Universe
`docs://esg-categories`	ESG Categories
`docs://ungc`	UNGC
`docs://llms.txt`	Structured index of the documentation

The server also exposes a dashboard://urls resource containing parameterized URL templates for company-level dashboards (ESG Controversies, Case Monitoring, SDG Positive Impact, Risk Events) and portfolio-level dashboards (ESG Portfolio Monitoring, Case Portfolio Monitoring, SDG Portfolio Monitoring).

Tools

Every v3 API route is registered as an MCP tool. Tool names are based on the HTTP method and path.

For example:

Route	Tool name
`GET /entities`	`GetEntities`
`GET /entities/{id}`	`GetEntity`
`GET /entities/{id}/events`	`GetEntityEvents`
`POST /universes`	`PostUniverses`

A subset of v2 routes is also available as tools for endpoints not yet covered by v3. These are marked as deprecated in their description and should be replaced by the v3 equivalent when one exists.

Client setup

When adding a new MCP server to your LLM provider, you'll have to use your own user account. Each request made by the llm will have the same permissions as the user account.

Claude Code

Run the following command to register the TextReveal MCP server:

claude mcp add --transport http textreveal-mcp https://mcp.textreveal.com/mcp

Then, inside a Claude Code session, run /mcp, select textreveal-mcp, and follow the authentication flow to authorize the server. Once authorized, the tools are available immediately in your session.

To verify the connection:

claude mcp list

You should see textreveal-mcp listed.

Troubleshooting

Authentication errors on tool calls

Claude Desktop / Claude.ai

In the Claude interface, select Customize in the sidebar, then Connectors -> Add custom connector.

Give it a custom name, then paste https://mcp.textreveal.com/mcp in the URL field. Click Add to save the connector.

For custom use-case, follow Claude's documentation.

A tool is missing

ChatGPT

To add a new App in ChatGPT, you need to have the Developer Mode enabled. To do so, open the Settings, then go on Apps -> Advanced settings, and enable Developer Mode.

Once enabled, you can click on Create app and fill the form with the following values:

Field	Value
Name	TextReveal MCP
MCP Server URL	https://mcp.textreveal.com/mcp
Authentication	OAuth

Then when starting a new conversation, click on the + icon and select More -> TextReveal MCP or by tagging it with @TextReveal MCP in your message.

More information can be found in the ChatGPT documentation.