Compile retiring OpenAI Custom GPT Actions into Cloudflare Remote MCP & SKILL.md ($5/$15).
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Compile an OpenAI Custom GPT OpenAPI 3.0/3.1 JSON or YAML Action schema into MCP tools/list definitions, SKILL.md, and Cloudflare Worker migration code before the Dec 11, 2026 retirement.
Raw OpenAPI 3.0/3.1 JSON or YAML string
Optional public URL to openapi.json or openapi.yaml
Optional Custom GPT system instructions for SKILL.md
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http action2mcp https://action2mcp.ak10082247.workers.dev/mcp
Compile an OpenAI Custom GPT OpenAPI 3.0/3.1 JSON or YAML Action schema into MCP tools/list definitions, SKILL.md, and Cloudflare Worker migration code before the Dec 11, 2026 retirement.
| Argument | Type | Description |
|---|---|---|
| schema | string |
| Raw OpenAPI 3.0/3.1 JSON or YAML string |
| auth_type | string |
| schema_url | string | Optional public URL to openapi.json or openapi.yaml |
| instructions | string | Optional Custom GPT system instructions for SKILL.md |
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"description": "Raw OpenAPI 3.0/3.1 JSON or YAML string"
},
"auth_type": {
"enum": [
"auto",
"none",
"bearer",
"api_key_header"
],
"type": "string",
"default": "auto"
},
"schema_url": {
"type": "string",
"description": "Optional public URL to openapi.json or openapi.yaml"
},
"instructions": {
"type": "string",
"description": "Optional Custom GPT system instructions for SKILL.md"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 19, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.