Connected AI · MCP
Your AI brings analytical capability. Innguma brings context.
Currently compatible with
ChatGPT
Claude
Microsoft Copilot
Gemini
Ask what has changed around a competitor, request a summary of what has been published this quarter on a technology or prepare a briefing for management. Through MCP, ChatGPT, Claude, Microsoft Copilot and Gemini can work with the knowledge and capabilities that Innguma makes available to each integration.
Innguma provides connectivity through the Model Context Protocol (MCP), the open standard these assistants use to connect to external systems.
ClaudeConnected to Innguma · MCPIllustrative example · simulated conversation
Practical use cases
What you can do with Innguma’s MCP.
From ChatGPT, Claude, Microsoft Copilot or Gemini, in natural language and on your own monitoring. Each organisation decides which capabilities it enables in its integration.
Query
- 01
Check your inbox
Ask directly about the signals that have reached Innguma and make any request about them: summarise, compare, prioritise or prepare a briefing.
Example request‘What came in this week about solid-state batteries? Summarise it in five points.’ - 02
Work through the Factory inbox
Read each signal in full with its source, check unread items by source and folder, mark signals as read, starred or pending, add the analyst’s comment and delete signals or empty the inbox.
Example request‘Mark today’s patents as pending and comment on the most relevant one for the team.’ - 03
Search what has been published
Search the content published in the Library and get answers with their source and date.
Example request‘Search what was published this year about green hydrogen and group it by country.’ - 04
Comment and rate in the Library
Read each document with its fields and tags, comment on it, rate it from 1 to 5 and see the team’s comments and rating.
Example request‘Which documents from this quarter have the best rating from the team?’
Configure
- 05
Add sources from the chat
RSS, web content, web capture, AI feed, newsletters, patents, YouTube, LinkedIn, Bluesky, Facebook, Telegram, Price Tracker, search engines and databases. The assistant asks for what each type needs, validates the source before creating it and shows you a sample of what it would bring in. You can then edit, move, share or delete it.
Example request‘Add this competitor’s website as a source and show me what it would bring in.’ - 06
Organise folders, catalogue and reminders
Create, rename, move and delete folders; search and follow sources in Innguma’s catalogue; refresh a source and adjust its publication options. And create reminders for whatever is worth reviewing regularly, such as a competitor’s website.
Example request‘Remind me every Monday to review this competitor’s pricing page.’ - 07
Adjust the filters
Change the filters of each watch by asking in natural language.
Example request‘Exclude Formula 1 news from the mobility watch.’ - 08
Audit the performance of your sources
Review which sources contribute and which generate noise, and get suggestions for improvement.
Example request‘Which sources haven’t contributed anything relevant in the last quarter?’
Publish
- 09
Publish automatically
Publish Factory signals to a category, with a summary, insight, tags or any additional field your organisation uses.
Example request‘Publish this signal with a summary, the insight for R&D and the right tags.’ - 10
Create news items on your corporate portal
Write and create a news item directly on your Innguma corporate portal, and feature it if you want it to appear first.
Example request‘Create a news item with this week’s summary on hydrogen and feature it.’ - 11
Manage what is published
Edit, publish or unpublish, delete and restore from the bin without leaving your assistant.
Example request‘Correct the headline of the signal published yesterday about the battery plant.’
Organisation
- 12
Check your organisation
Check your identity and permissions, who the analysts and readers in the organisation are, and the totals of readers, analysts and news items.
Example request‘How many readers do we have and how many news items have we published this month?’ - 13
Organise the Library folders
Create, rename, move and reorder the Library folders, delete the empty ones and decide which groups can see each one.
Example request‘Create a Hydrogen folder inside Energy that only the R&D group can see.’ - 14
Maintain the tag tree
Add, rename and move tags within the tree, mark the ones no longer in use as obsolete and delete branches after seeing which news items they affect.
Example request‘Move the Electrolysers tag under Hydrogen and mark Alkaline cells as obsolete.’
AI inside Innguma
Even if you do not connect any assistant, AI is already working for you.
It filters, groups duplicates and classifies each signal before it reaches your team, so your analysts start with what matters.
Frequently asked questions
What we are usually asked.
What is MCP?
Model Context Protocol is an open standard that defines how an AI assistant connects to external systems. Innguma offers it so that ChatGPT, Claude, Microsoft Copilot and Gemini can work with its knowledge.
I want more informationWhich assistants are compatible?
ChatGPT, Claude, Microsoft Copilot and Gemini. These are the assistants with which Innguma's MCP connectivity is currently compatible.
I want more informationDoes the AI see all the information?
It works with the knowledge and capabilities that Innguma makes available to each integration. That scope is defined with your organisation.
I want more informationCan it publish or change things on its own?
AI queries, summarises and proposes. The capabilities available depend on what Innguma enables in each integration, and people decide what is published and who receives it.
I want more informationIs it available?
Yes. Innguma offers connectivity through MCP, currently compatible with ChatGPT, Claude, Microsoft Copilot and Google Gemini.
I want more informationConnected AI · MCP
Watch your assistant work on your own monitoring.
We show you how ChatGPT, Claude, Microsoft Copilot or Gemini answer, through MCP, with what your team has already captured and validated.