Skip to contentandromesh labs

Services

Custom MCP development services

Three ways to work with us, all with published starting prices. Whether you're shipping a connector for your product or connecting agents to internal systems, the craft is the same.

Every engagement is a fixed-price proposal after a 30-minute scoping call.

MCP Audit & Eval

from $4,500 1–2 weeks

We run your existing or generated MCP server through the same evals and security review we use on our own builds.

What’s included

  • Tool-selection eval scorecard across Claude, ChatGPT, and Cursor
  • Context-budget analysis of your tools/list
  • Auth review: OAuth 2.1 flow, scopes, and token handling
  • Prompt-injection and data-exfiltration threat review
  • Prioritized fix list with redesigned tool signatures

Best for: Teams whose server works in the Inspector but disappoints in real clients.

Custom MCP Build

from $18,000 Typically 3–6 weeks

A production MCP server designed around the tasks your users and their agents actually perform. Secured, evaluated, and handed over.

What’s included

  • Scoping and tool-design workshop
  • Remote MCP server (Streamable HTTP) in your repository
  • OAuth 2.1 with per-user, least-privilege scopes
  • Human approval gates and audit logging for consequential actions
  • MCP Apps UI where a widget beats a wall of text
  • Eval suite and scorecard, wired into your CI
  • Deployment to your cloud, runbook, and team walkthrough
  • Claude connector and ChatGPT app submission prep

Best for: SaaS teams shipping a connector, and enterprises giving agents governed access to internal systems.

Care & Evolve

from $1,500/mo Ongoing, monthly

MCP moves fast. We keep your server current with the spec and the clients, and keep the evals green.

What’s included

  • Spec and client-compatibility tracking
  • Eval regression runs on every change
  • New tools and tool refinements
  • Monitoring, alerting, and incident support

Best for: Anyone whose MCP server is now part of the product.

What every build includes

  • Task-shaped tool design, reviewed with you before we write code
  • Remote MCP over Streamable HTTP (stdio where it fits)
  • OAuth 2.1, per-user tokens, least-privilege scopes
  • Structured errors, pagination, and idempotent writes
  • Approval gates and audit logs on consequential actions
  • Eval suite across Claude, ChatGPT, and Cursor, in your CI
  • Your repo, your cloud, documentation, and a runbook

Go deeper: how we design MCP tools and how we secure remote MCP servers.

Not sure which one fits?

Tell us what you have and what agents should be able to do. We’ll recommend the smallest engagement that gets you there.

Start an MCP project