Export & Integration

Once you’ve built and tested your MCP manifest in ContextMesh, it’s time to take it live. ContextMesh makes it effortless to package, integrate, and share your workflow across any environment that supports the Model Context Protocol.

1. One-Click manifest.json Export

With a single click, export your entire workflow as a standards-compliant manifest.json. This JSON bundle contains:

  • Metadata: name, version, author, description, tags

  • Nodes: all Resources, Prompts, Tools, and Sampling steps

  • Connectors: data-flow edges linking nodes

  • Capability declarations: MCP capabilities required by your workflow

How to export

  1. In Manifest Studio, click Publish → Export Manifest

  2. Download the generated manifest.json

  3. Drop it into any MCP-enabled host or client

// manifest.json (excerpt)
{
  "manifestName": "My Solscan Connector",
  "version": "1.0.0",
  "author": "Acme Corp",
  "tags": ["solana", "blockchain", "explorer"],
  "nodes": [
    { "id": "resource-1", "type": "Resource", "uriTemplate": "https://api.solscan.io/account/{address}" },
    { "id": "prompt-1",   "type": "Prompt",   "name": "AnalyzeAccount", "arguments": ["address"] },
    { "id": "tool-1",     "type": "Tool",     "name": "FetchBalance",   "inputSchema": { "address": "string" } }
  ],
  "connectors": [
    { "from": "resource-1", "to": "prompt-1", "mapping": { "address": "address" } },
    { "from": "prompt-1",   "to": "tool-1",   "mapping": { "address": "address" } }
  ]
}

2. SDK Generation (React, Python, Node.js)

Automatically generate client libraries in your favorite language. ContextMesh introspects your manifest and produces:

  • Type-safe functions to invoke each Resource, Prompt, Tool, or Sampling step

  • Configuration helpers for MCP endpoint URLs, session IDs, and credential injection

  • Full docs and IntelliSense support for arguments, return types, and error handling

Supported runtimes

  • React Hooks & Context Provider

  • Python package (pip install contextmesh-sdk)

  • Node.js module (npm install @contextmesh/sdk)

# Example: generate a Python SDK
cm export-sdk --lang python --manifest ./manifest.json
# client.py (generated)
from contextmesh import ContextMeshClient

client = ContextMeshClient(endpoint="https://mcp.myapp.com", api_key="…")

# Call the Solscan resource
account_data = client.resources.get_account(address="H3yZ…")
analysis     = client.prompts.analyze_account(address="H3yZ…")
balance      = client.tools.fetch_balance(address="H3yZ…")

3. Shareable Links & Embeds

Turn any manifest into a live, interactive widget that can be embedded or shared:

  • Public Demo Links: Host your manifest on ContextMesh Cloud and share a 🔗 URL

  • Iframe Embeds: Drop a snippet into docs or dashboards for inline interactive previews

  • GitBook Blocks: Include live “Try it now” panels directly within your GitBook documentation

<!-- Example: iframe embed -->
<iframe
  src="https://app.contextmesh.com/embed?manifestId=abc123"
  width="100%"
  height="600"
  frameborder="0"
  allow="clipboard-write"
></iframe>

Pro tip: Use URL parameters to pre-fill inputs, control read/write modes, and capture results via webhooks.

By combining one-click manifest export, multi-language SDKs, and seamless embeds, ContextMesh empowers you to build once and deploy everywhere. Your MCP-powered workflows become instantly portable, shareable, and extensible - across teams, apps, and cloud platforms.

PreviousCollaboration & Version ControlNextGetting Started

Last updated