NEWThe Value Intelligence Buyer’s Guide 2026.Read the guide →

A value harness for the AI your team already uses

Every agent has a harness. valueIQ MCP gives yours the part that knows what your product is worth: connect Claude, ChatGPT, Cursor or your own agents, and they build business cases on your value models, coach deals and pull competitor pricing without opening another tab.

2,000+
Pricing pages analysed
500+
Value models generated
100+
Pricing engagements
15+
Years of pricing expertise
The problem

Your AI can write. Can it reason about value?

A general-purpose assistant will write a fluent business case in seconds. The trouble starts when finance asks where the numbers came from.

  • The figures come from the prompt and the model's general knowledge, so nobody can trace them to a source.
  • Each person pastes different context, so two people get two answers for the same deal.
  • Assumptions are guessed, and the customer never confirms them.
  • Every new workflow starts from scratch, with no shared model of how your product creates value.
A prompt alone produces an untraceable figure; valueIQ produces the same figure with its source shown.

Illustrative example.

The value harness

Every agent has a harness. Yours needs a value harness.

A value harness is the part of an AI agent's harness that knows what a product is worth to a specific buyer, and can show its working. The value harness is how value intelligence reaches an AI agent.

A value model
Value drivers, the equation behind each, the variables that feed it, and sourced benchmarks. In valueIQ today: generated for your product from its URL in about 25 minutes.
Deal context
The RFP, discovery notes, call transcripts and the buyer's own figures. In valueIQ today: attach documents and notes to a deal, and the business case is fitted to that buyer.
Confirmation and provenance
Who set each number (a market default, the seller or the buyer) and whether the buyer confirmed it. In valueIQ today: every input is marked, and buyers confirm their own inputs on a shared business case.
Identity and permissions
The agent works as its user, with that user's role. In valueIQ today: valueIQ MCP runs under your own identity and workspace permissions.
Memory across tools
The deal's conversation and working state stay with the deal. In valueIQ today: session checkpoints, and a thread started in Claude continues in the valueIQ app.
One record after the signature
Measured results against the business case's own equations. In valueIQ today: realized value tracked with the same equations, for QBRs, renewals and expansion.

Other harnesses govern what an agent does. A value harness governs what it knows about value. The value model is the part the others depend on.

What is a value harness? Read the post
What your AI gets

What your AI gets that a prompt can't give it

A value model generated for your product.

valueIQ generates a value model for your product from its URL in about 25 minutes: the value drivers, the equation behind each one, the variables that feed it, and cited market benchmarks. It is ready for deals straight away, and your AI reads it through valueIQ MCP.

Competitor value models and pricing analysis on the same deal.

Attach a competitor's value model to a deal and compare it with yours driver by driver. Pull competitor pricing analysis from 2,000+ analyzed pricing pages. The comparison is about what each option is worth to this buyer.

A business case built from both.

Your AI gets an executive-ready business case in which every number shows whether it is a market default, a figure the seller set or a number the customer confirmed.

Consistent outputs across people and tools. Numbers you can inspect. Less pasted context. You decide what is sent.

What your assistant can do

Three jobs, one connection

01

Coach on live deals

Ask about a real opportunity by name. Get methodology-grounded next steps, risk calls, and executive-ready framing, scoped to that deal's context.

02

Pull value context

Business cases, driver assumptions, pricing intelligence, and prior estimates flow into the conversation as structured, cited context.

03

Your identity, your scope

Every request runs under your own identity and workspace permissions, so your assistant sees exactly what you see.

What's live today

Read and understand

  • Live
    Read Your Deals and Business Cases

    Ask for any deal by company name and get the value breakdown, each driver's calculated impact, the risks valueIQ sees in the deal, and the readiness assessment from the latest value story.

    Try asking Pull up the Acme deal and walk me through the value breakdown.

  • Live
    Browse Workspaces, Products, and Market Benchmarks

    See which products are ready for deals, read each product's value drivers and equations, check the low, typical and high market figures for a variable, and list deals by stage.

    Try asking Show me the value drivers and equations for our core product.

  • Live
    Read Pipeline Rollups and Credit Balance

    Get the workspace's pipeline rollup, with active deal count, total expected value and a per-stage breakdown, and check how many credits the workspace has left. You can also list earlier coaching threads on a deal.

    Try asking How many credits do we have left?

Generate and share

  • LiveUses credits
    Generate a Business Case for a Deal

    Ask for a business case for a company and product, and valueIQ creates the deal and generates the customer model. You can also ask for an executive briefing or a sales deck. Generating draws credits.

    Try asking Build a business case for Acme using our platform.

  • BetaUses credits
    Create Products and Edit Value Models, with Undo (Beta)

    Create a product from a company website and a product description, then edit its value model: rename drivers, mark them not applicable, or add notes. The last edit can be undone on request, and you can see the edit history and move the product to any point in it. Archive a product and restore it later; the assistant lists the variants and deals that go with it and waits for your confirmation. Creating a product draws credits.

    Try asking Create a product for Acme using their website.

  • Live
    Manage Value Stories, Comments, and Customer Invites

    List and open a deal's value stories, choose the top drivers and lock a story before a presentation, and read and reply to the customer's comments. Edit a headline or paragraph, switch the currency, show or hide a box, archive a story, and bookmark or restore any earlier version. You can also see who has been invited to a story and whether they've claimed it.

    Try asking What comments has the customer left on the story?

Coach and compete

  • Live
    Run Deal Coaching Grounded in the Deal's Business Case

    Describe an objection or a stalled moment, such as price pushback or "good enough", and get a response backed by the deal's quantified value drivers. Follow-up drafts come back for you to review, and nothing is sent for you.

    Try asking They pushed back on price. Give me a response backed by the drivers.

  • Live
    Pull Competitive Pricing Analysis

    Get structured data from a competitor's published pricing page: tiers, features, billing model and usage metrics, with competitive positioning. Analyses are cached for 90 days, so the data is current but not real time.

    Try asking Pull the competitive pricing analysis for the Acme deal.

  • Live
    Add a Competitor Value Model to a Deal

    Attach a competitor's value model to a deal and get a side-by-side, driver-by-driver comparison with your own. The competitor's model must already exist in your workspace.

    Try asking Add Globex to the Acme deal as a competitor.

Manage and connect

  • LiveUses credits
    Create and Update Deals

    Create a deal, update customer variables such as headcount and recalculate, and log notes and discovery details. Archive deals you've lost and restore them when you need them.

    Try asking Update their headcount to 1,200 and recalculate.

  • Live
    Attach Documents to a Deal

    Upload a file such as a discovery call transcript to a deal, so coaching and the customer model can use it. You can list a deal's documents and stop one being used in coaching.

    Try asking Attach this discovery call transcript to the Acme deal.

  • Set up with our team
    Link Deals and Products to Your CRM

    Store a CRM record reference on a deal or product, and read the deal's CRM scorecard: Value Capture, zone, estimated value and the value summary link the CRM displays. Salesforce, HubSpot and SugarCRM connectors are available today, and our team can help you set one up.

    Try asking What is the CRM scorecard for the Acme deal?

  • Live
    Resume Work Across Clients with Session Checkpoints

    Save where a coaching session or multi-step task has got to, then pick it up later or from another client, such as moving from Claude Desktop to Cursor, instead of starting again.

    Try asking Save a checkpoint of this session.

Example prompts are illustrative. Swap in your own deal, company and product names.

How it fits

How the pieces fit: a Skill, MCP and valueIQ

An agent is a model plus a harness. The Skill explains how to do the work. MCP gives your AI access to the systems that do it. valueIQ is the part of the harness that knows what your product is worth.

Generative AI
Writes and reasons from the context it is given.
A Skill
Instructions for how the work is done: the steps, the questions to ask and the review rules.
MCP
An open standard that connects your AI to the systems that do the work, with the permissions you already have.
valueIQ
A value harness for your AI: the value models, business cases, competitor value models and pricing analysis your AI calls.
  • Use a Skill alone when the work is a procedure.
  • Use MCP alone when your AI only needs access.
  • Use both when your AI needs to follow a method and call real intelligence.
Workflows

Start with one workflow

Deal value agent

An agent in Claude reads a deal, asks for the inputs that are missing (no more than five questions), generates the business case, and drafts a follow-up for you to review. Nothing is sent without you.

Value story Skill

A Skill that turns an approved business case into the format each stakeholder needs: a sales deck, a one-pager, an executive briefing, a discovery guide or a co-session discovery guide.

Competitive value response

When a buyer says a competitor is cheaper, your agent pulls the competitor's pricing analysis and compares the two value models driver by driver, then drafts a response for you to review. The competitor's value model must already be in the workspace.

Value realization planning

A planning workflow your team runs in its own AI over the approved business case, to agree what will be measured after signature. Your team then records results in valueIQ, which tracks realized value against the business case with the same equations. Results flowing in automatically from your own systems are coming soon.

Getting started

Seven steps to your first workflow

  1. Choose one workflow.
  2. Decide which source of truth the AI should use.
  3. Connect valueIQ MCP.
  4. Add your instructions or a Skill.
  5. Test it on a real deal.
  6. Set the review rules for what can be sent.
  7. Measure whether the first workflow saves time or improves the case.

Connect guides: Claude · ChatGPT · Cursor · VS Code Copilot · Windsurf

Security

Security and control

  1. Your AI drafts
  2. You review
  3. You send

Nothing sends on its own.

OAuth 2.1 + PKCE
Your AI connects with no long-lived keys to manage.
Per-request JWT
Each request gets its own short-lived token.
Workspace-scoped isolation
Data is isolated by workspace with row-level security.
Input validation
Inputs are validated before they run.
Audit log
Every call is logged. Enterprise customers can read the audit log through the API.
Open standard

Built on the open MCP standard

Tools

Actions the assistant can invoke against valueIQ.

value_model_get, deal_coaching_get_response

Resources

Addressable context the assistant can read on connect.

valueiq://deals/{id}/model

Prompts

Named, versioned skills the assistant can adopt.

valueiq://skill/objection-handling
Cost

Free to connect

Connecting valueIQ MCP is free. Reading deals, coaching and drafting are free; creating a deal or generating a deck uses credits from your valueIQ Sales plan.

MCP scope by plan

Free

  • 500 credits a month
  • Core deal and business case access

Professional

  • Full deal management
  • Business case generation
  • Coaching

Teams

  • Pipeline health dashboard
  • Shared workspace context
Roadmap

What's coming next

Available today

CRM Bidirectional

Salesforce, HubSpot and SugarCRM connectors are available today, so coaching outcomes can flow back into your pipeline of record. Our team can help you set one up.

Coming soon

A2A Agent Cards

Agent-to-agent discovery so other agents can call valueIQ as a peer.

Coming soon

Marketplace Listings

Smithery, Glama, and Databricks: one-click installs from the surfaces your team already trusts.

FAQ

Questions about valueIQ MCP

MCP (Model Context Protocol) is an open standard for connecting AI tools to live external data. Released by Anthropic in late 2024 and adopted by OpenAI, Google, and Microsoft in early 2025. Before MCP, using AI for deal work meant copying notes into a prompt and hoping the output fit your situation; with MCP, your AI tool connects to your data once and has the deal, the value context and the competitive context available in every conversation.

Build the first workflow with us.

Bring one workflow you want your AI to handle. We will map it to the tools and a Skill with you in a working session.