MCP & AGENT INTEGRATIONS
Connect the selected SpinForge products to compatible AI clients, coding tools and agent workflows through their available MCP interfaces, with CLI and headless alternatives.
Available products. A customer-specific workflow.
CLI · MCP · Headless
AI-platform companies, engineering teams and integrators adding reliable computational tools to an existing client or assistant.
A useful implementation has a specific job. The scope is built around your original requirements, environment and acceptance criteria.
A natural-language answer is not a substitute for executing a supported mathematical or infrastructure operation.
Expose the relevant product operations to a compatible client and return the actual tool outputs for interpretation. The computational result comes from the product.
Users should not need to copy every request into another graphical application just to use a licensed capability.
Use the available MCP interface from the chosen client, or invoke the CLI headlessly from existing scripts. Keep the customer workflow in its current environment.
Discovery, credentials, expensive operations and result handling need a defined integration rather than an unrestricted connection.
Scope permitted operations, transport, credentials, budgets and approvals. Document the supported client setup and handling of failed or incomplete operations.
The final order identifies the selected products, integration work and operating responsibilities. Included components and work are not charged again.
The products and exposed operations that belong in the customer’s workflow, without an unnecessary full-catalog connection.
Product-specific CLI, MCP and headless access information, with compatible client, transport and authentication details reviewed for the deployment.
A defined way to preserve status, evidence and errors when returning results to the AI client.
An optional client configuration, acceptance test and support handoff. Public directory submissions are separate from private use.
A team uses a compatible AI client to request a supported model check and summarize its outcome. The integration keeps the original artifact identity and does not turn an unproved result into a proof.
Explanation only—not a live product run or a claimed customer deployment.Select the product operations, client environment and least-privilege access required for the task.
Agree the supported models, execution environment and acceptance measures before committing to the delivery.
Choose the relevant capability.
Not every workflow needs every product.
CLI, MCP and headless are included in the relevant product entitlement—not three separate subscriptions. Guided onboarding is $5,000 for its existing defined scope; a custom agent or directory integration is separately quoted.
Yes. CLI, MCP and headless access are available across all 64 named products. Request the selected product’s connection details for the client you use.
This page does not claim a directory listing or approval. Direct supported integration and any third-party publishing process are different activities.
No. Product licensing, credentials, allowed operations, spending limits and customer approval rules still apply.
UNISOLVE exposes solve_any through a stdio MCP server. It detects supported files, dispatches or races compatible profiles and gates acceptance through solver-specific and generic checks.
Commission the model, integration and acceptance work needed for your task. Existing products stay independently available.
One capability. A connected workflow.
A team to put it into practice.