One hosted connection for your AI tools

A planned hosted MCP service will provide a shared remote entry point for compatible clients. Keep authentication and access controls on a trusted server; start validating your workflow with the public HTTP API today.

Hosted MCP is planned. Use the public HTTP API today.

  • Reduce repeated local setup and maintenance
  • Validate credentials and access rules on a trusted server
  • Combine tool access with reusable Skill workflows

For authorized web access, development, testing and data workflows.

HTTP API requests and session management illustration

Clear connections and clear responsibilities

Start with the smallest usable path, then add tools and workflow complexity as needed.

Call the public API

Use your existing HTTP client and control requests, sessions and result handling. Available today.

API documentation

Organize tasks with a Skill

Give the agent access rules, then prepare parameters, check status and handle failures by task.

Cloudbypass Skill

Connect through MCP

Existing HTTP tools need custom headers and secure credential injection. The hosted service is still planned.

Hosted MCP

From API validation to a hosted connection

Prove the workflow with one inspectable request before adding complex configuration.

  1. Validate the public API first

    Use one authorized target to confirm versions, proxies and responses. Keep a minimal reproducible request.

  2. Configure after release

    Wait for the published endpoint and authentication guide. A remote service manages its own credentials rather than automatically inheriting local variables.

  3. Check tool and target results

    Once released, validate tool discovery, parameters, errors and content, including access permissions and stop conditions.

Use the public API while hosted MCP is planned

Examples use minimal parameters and no real key. Follow the documentation for V2 / V2s proxy and session settings.

Start with a minimal requestHTTP API
curl --include --request GET \
  "https://api.cloudbypass.com/" \
  --header "x-cb-host: example.com" \
  --header "x-cb-apikey: ${CLOUDBYPASS_APIKEY:?}" \
  --max-time 90

# POSIX shell example; set the key securely first.
# Check x-cb-status, target status and content.
# For V2 / V2s, add version and stable proxy parameters.
Set credentials in the runtime first. Examples do not run on this page.
HTTP API requests and session management illustration

What MCP and a Skill each do

MCP connects tools; a Skill describes the workflow. They can work together without replacing target permissions or business rules.

Proxy network and access routing illustration
ResponsibilityHosted MCP (planned)Skill / project workflow
ConnectionExpose a remote tool entry point and access controlsDefine task steps and failure conditions
CredentialsInject and validate securely in the actual serviceReference configuration without storing real secrets
Site logicReturn status, errors and actual responsesHandle pagination, extraction, deduplication and storage
AvailabilityHosted service planned; no endpoint published hereStart with the guide and published OpenClaw Skill

Put your integration to work

Choose a task guide for access strategies, implementation steps and troubleshooting.

Planning and troubleshooting

Development and tool integration

Business use cases

Explore all 14 use cases

Frequently asked questions

Can I connect to the hosted MCP service now?

Not yet. This page explains the planned service, not a live endpoint, installer or tool contract. Use the public HTTP API today or a reviewed HTTP MCP tool with custom header support.

Can every target be accessed?

No universal guarantee is made. Relevant access capabilities include Cloudflare JavaScript Challenge, Turnstile, DataDome and Imperva / Incapsula, subject to target and runtime validation. This integration does not advertise reCAPTCHA or hCaptcha support or replace account permissions.

Where should API keys and proxy passwords go?

Keep them in secure configuration in the actual runtime or a trusted server. Do not put them in browser code, shared repositories, prompts, screenshots or public logs. Remote services have separate environments.

Do MCP or Skills automatically handle pagination and extraction?

Not by default. The access layer handles requests and responses; workflows or site recipes handle pagination, field extraction, deduplication, login state and business-specific signatures. Diagnose failures instead of retrying indefinitely.

Validate one request before scaling up

Confirm target compatibility, then bring the workflow into your project.