AI tools are getting better at answering questions. The problem is they can only answer questions using information they already have. Ask an AI assistant about your portfolio's Scope 2 emissions and it will either guess, tell you it doesn't know, or ask you to paste the numbers in yourself.
MCP changes that.
The missing piece between AI and your data
MCP (Model Context Protocol) is an open standard that lets AI assistants connect directly to external platforms and query live data in real time. Think of it as a structured bridge: instead of copying figures out of a system and into a prompt, your AI tool can call the source directly and get accurate, up-to-date answers.
It was developed by Anthropic and released as an open standard in late 2024, and it is now supported by a growing range of AI tools including Claude, Microsoft Copilot, and others.
The practical effect is significant. Rather than using AI to process data you have manually collected and supplied, you can connect your AI assistant to the systems that hold your verified data, and let it query them directly.
What this means for ESG teams
Sustainability reporting is a data-heavy discipline. ESG leads, fund managers, and sustainability officers spend significant time pulling figures from platforms, compiling spreadsheets, and formatting outputs for LP reports, regulatory submissions, board packs, and investor/vendor questionnaires.
Most of that work is not insight generation. It is data retrieval and reformatting - tasks AI is well-placed to automate, but only if it has access to the right data.
That is the gap MCP closes.
What you can do with KEY ESG's MCP server
KEY ESG's MCP server gives AI assistants read access to your live ESG data. Once connected, your AI tool, whether that's Claude, Copilot, or another MCP-compatible assistant, can query your metrics, targets, action plans, policies, and portfolio performance in real time.
Here is what that looks like in practice.
LP and investor reporting
Your AI assistant can generate a full portfolio ESG performance summary, including Scope 1 and 2 emissions, gender diversity metrics, target progress, coverage gaps, in a single session. Rather than spending hours consolidating spreadsheets from portfolio companies, a sustainability lead can ask a question and receive a structured answer grounded in verified platform data.
ESG questionnaire completion
Procurement portals, vendor questionnaires, sustainability-linked finance applications, customer RFPs, and regulatory framework submissions all ask overlapping ESG questions. An AI connected to KEY ESG can answer them in real time, pulling live metric values, policy details, and target progress, and even work through an Excel-format questionnaire directly. No manual data extraction required.
Portfolio or entity health checks
Sustainability managers can ask: "Where do I have gaps in my Scope 2 data this year? Which targets are at risk across the portfolio/company?" The MCP server returns structured answers with coverage counts and status flags - an instant health check, without logging into the platform.
Board and exec briefings
Before a board meeting, a CFO or CEO can ask their AI assistant for an on-demand ESG summary: key metrics compared to the prior year, target progress, and current improvement initiatives. The briefing is composed from live KEY ESG data, in seconds, without involving the ESG team.
Carbon performance tracking
"What are our Scope 1, 2, and 3 emissions this year, and are we on track for our 2030 net zero target?" - A complete carbon performance summary, with trajectory context, available as a prompt.
Policy audits
"Which ESG policies haven't been reviewed in the last 12 months?" - The MCP server returns all policies with review dates and document download links. An AI can flag overdue reviews and provide direct access to the documents.
Who it is for
The MCP server is useful for anyone who regularly needs ESG data but does not want to log into a platform to get it.
For solo ESG leads at PE firms managing 15 to 25 portfolio companies, it means LP report preparation that used to take days can be reduced to a focused session. For fund managers fielding investor questions, it means self-service answers without waiting for a report to be compiled. For corporates managing multi-framework reporting obligations, it means ESG questionnaires and covenant disclosures can be answered directly from live, auditable data.
The connector is read-only, it does not make changes to your KEY ESG account or data, and uses OAuth authentication, so your credentials are never shared with the AI tool directly.
Why this matters now
ESG reporting obligations are increasing; CSRD, SFDR, EDCI, and a growing range of voluntary frameworks, at the same time as the teams responsible for them are under pressure to do more with less. AI tools offer genuine leverage, but only when they have access to the right data.
An AI assistant working from last year's PDF report or a manually pasted spreadsheet will produce answers that are out of date, hard to verify, and time-consuming to compile. An AI assistant connected to a live, structured data platform produces answers that are accurate, traceable, and available on demand.
That is the shift MCP makes possible - and it is available now, in beta, for KEY ESG customers.
Getting started
The KEY ESG MCP server is available in beta for Claude. If you have questions about the connector or would like to see it in action, get in touch with our team.
AI tools are getting better at answering questions. The problem is they can only answer questions using information they already have. Ask an AI assistant about your portfolio's Scope 2 emissions and it will either guess, tell you it doesn't know, or ask you to paste the numbers in yourself.
MCP changes that.
The missing piece between AI and your data
MCP (Model Context Protocol) is an open standard that lets AI assistants connect directly to external platforms and query live data in real time. Think of it as a structured bridge: instead of copying figures out of a system and into a prompt, your AI tool can call the source directly and get accurate, up-to-date answers.
It was developed by Anthropic and released as an open standard in late 2024, and it is now supported by a growing range of AI tools including Claude, Microsoft Copilot, and others.
The practical effect is significant. Rather than using AI to process data you have manually collected and supplied, you can connect your AI assistant to the systems that hold your verified data, and let it query them directly.
What this means for ESG teams
Sustainability reporting is a data-heavy discipline. ESG leads, fund managers, and sustainability officers spend significant time pulling figures from platforms, compiling spreadsheets, and formatting outputs for LP reports, regulatory submissions, board packs, and investor/vendor questionnaires.
Most of that work is not insight generation. It is data retrieval and reformatting - tasks AI is well-placed to automate, but only if it has access to the right data.
That is the gap MCP closes.
What you can do with KEY ESG's MCP server
KEY ESG's MCP server gives AI assistants read access to your live ESG data. Once connected, your AI tool, whether that's Claude, Copilot, or another MCP-compatible assistant, can query your metrics, targets, action plans, policies, and portfolio performance in real time.
Here is what that looks like in practice.
LP and investor reporting
Your AI assistant can generate a full portfolio ESG performance summary, including Scope 1 and 2 emissions, gender diversity metrics, target progress, coverage gaps, in a single session. Rather than spending hours consolidating spreadsheets from portfolio companies, a sustainability lead can ask a question and receive a structured answer grounded in verified platform data.
ESG questionnaire completion
Procurement portals, vendor questionnaires, sustainability-linked finance applications, customer RFPs, and regulatory framework submissions all ask overlapping ESG questions. An AI connected to KEY ESG can answer them in real time, pulling live metric values, policy details, and target progress, and even work through an Excel-format questionnaire directly. No manual data extraction required.
Portfolio or entity health checks
Sustainability managers can ask: "Where do I have gaps in my Scope 2 data this year? Which targets are at risk across the portfolio/company?" The MCP server returns structured answers with coverage counts and status flags - an instant health check, without logging into the platform.
Board and exec briefings
Before a board meeting, a CFO or CEO can ask their AI assistant for an on-demand ESG summary: key metrics compared to the prior year, target progress, and current improvement initiatives. The briefing is composed from live KEY ESG data, in seconds, without involving the ESG team.
Carbon performance tracking
"What are our Scope 1, 2, and 3 emissions this year, and are we on track for our 2030 net zero target?" - A complete carbon performance summary, with trajectory context, available as a prompt.
Policy audits
"Which ESG policies haven't been reviewed in the last 12 months?" - The MCP server returns all policies with review dates and document download links. An AI can flag overdue reviews and provide direct access to the documents.
Who it is for
The MCP server is useful for anyone who regularly needs ESG data but does not want to log into a platform to get it.
For solo ESG leads at PE firms managing 15 to 25 portfolio companies, it means LP report preparation that used to take days can be reduced to a focused session. For fund managers fielding investor questions, it means self-service answers without waiting for a report to be compiled. For corporates managing multi-framework reporting obligations, it means ESG questionnaires and covenant disclosures can be answered directly from live, auditable data.
The connector is read-only, it does not make changes to your KEY ESG account or data, and uses OAuth authentication, so your credentials are never shared with the AI tool directly.
Why this matters now
ESG reporting obligations are increasing; CSRD, SFDR, EDCI, and a growing range of voluntary frameworks, at the same time as the teams responsible for them are under pressure to do more with less. AI tools offer genuine leverage, but only when they have access to the right data.
An AI assistant working from last year's PDF report or a manually pasted spreadsheet will produce answers that are out of date, hard to verify, and time-consuming to compile. An AI assistant connected to a live, structured data platform produces answers that are accurate, traceable, and available on demand.
That is the shift MCP makes possible - and it is available now, in beta, for KEY ESG customers.
Getting started
The KEY ESG MCP server is available in beta for Claude. If you have questions about the connector or would like to see it in action, get in touch with our team.
