AI plugin (MCP) for assessment insight

One assessment, one source

A person completes a behavioral assessment, rated 27 out of 30 in review by the British Psychological Society. That profile is the only thing the AI plugin draws on.

Our platform provides the grounding

Our platform supplies the grounded data from that person's profile: scores, norms, bands and approved interpretation. Your own AI assistant writes the wording, so the phrasing can read a little differently from one answer to the next. What it is grounded in does not change.

Nothing beyond what they could already see

The plugin returns only what that person's own permission level allows, rechecked on every request. It is read only, so it cannot create, change or delete anything in your data, and there is no privileged back channel for a prompt to reach.

Works with the assistant you already use

The AI plugin is built on MCP, the open standard leading AI assistants now support. Nothing new for your people to log into, and nothing new for your IT team to run.

Employee

Manager

Power user

Can see: their own profile in full, a colleague's top two team roles and a team role comparison, and group insights for their own team and department

Can ask: anything about how they work, and what their manager is able to see about them

Cannot see: a colleague's full profile

Can see: everything an employee can, plus their team list, team level insights and each direct report's full profile

Can ask: how to approach a conversation, or how two people in the team work together

Cannot see: the full profile of anyone who is not their direct report

Can see: aggregated insights across the organization, the employee directory and organization structure, and usage analytics

Can ask: where capability sits and where the gaps are

Cannot see: individual profiles beyond their own employee or manager access

We do not train AI on your people

Your data is not used to train or fine tune any AI model, ours or your assistant's. We supply grounded data only. The wording comes from your own assistant, outside our system.

Certified to ISO/IEC 27001:2022

We are ISO/IEC 27001:2022 certified and operate a mature information security management system. The plugin is hosted on the same infrastructure as the rest of the platform.

No new subprocessors

The plugin runs on the same Microsoft Azure infrastructure as the rest of the platform, so it adds no one to your subprocessor list. We never receive your prompts or conversations. They stay with the assistant you have already approved.

Encrypted in transit and at rest

Data is encrypted in transit with TLS 1.2 or higher, and encrypted at rest.

Where your data goes

Hosted on Microsoft Azure in the EU, in Ireland and the Netherlands. The plugin passes data through without storing it. Security and audit logs are kept for 24 months, in line with the rest of the platform.

Frequently asked questions

If it is not here, ask us. You will get a fast answer from an expert, not a form reply.

Talk to us

Which AI assistants does the plugin work with?

Compatible AI assistants that support MCP, the open standard the plugin is built on, such as Copilot, Claude and ChatGPT.

What happens when someone joins or leaves?

Access is checked on every request, not set once. Identity is confirmed at our gateway, and license, entitlement and scope are rechecked before any data is selected. When someone's access changes in the platform, the plugin reflects it on their next request. Nobody maintains a second list.

Can an employee see who has looked at their insight?

Not today. What every employee can do is see their own profile in full, along with a colleague's top two team roles, a team role comparison and group insights for their own team and department. They can also ask the plugin directly what their manager is able to see about them.

Can a manager use this in a performance or disciplinary conversation?

It is built for development conversations, not employment decisions. It never gives a ranking, recommendation or verdict. Every response carries usage constraints stating it is not a sole basis for an employment decision, and requests such as "who should I promote or manage out" are declined and steered back to development. A person on your side always leads interpretation.

We are open about the limit: we never receive the prompt, and no protocol can compel a language model. We make guessing unnecessary rather than impossible. And a profile describes how someone works, not how well.

Put it in front of your security team

Tell us what your reviewers need and we will bring the right people to the call, including the documents that do not sit on a web page.

How the AI plugin works inside your current tools

How the AI plugin builds an answer

One assessment, one source

Our platform provides the grounding

Nothing beyond what they could already see

Works with the assistant you already use

Who sees what

Four measures, not one verdict

Someone with no completed assessment

Anyone beyond your access

A group that is too small

Who is the best of these people

The answers your security team asks for