Agentcore Register logo

Agentcore Register

Organization
agentic-community
agentcore-register

Given an MCP server URL, probe the server via curl to discover its metadata and tools, then generate a markdown file with copy-pasteable content for each field in the Amazon Bedrock AgentCore "Create record" form.

Overview

Publisheragentic-community
Repositorymcp-gateway-registry
Skill nameagentcore-register
Stars
969
Forks
244
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by agentic-community on GitHub. Read the source before you install it.

Installation

Install the Agentcore Register AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/agentic-community/mcp-gateway-registry/tree/main/.claude/skills/agentcore-register
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/agentic-community/mcp-gateway-registry.git /tmp/mcp-gateway-registry
mkdir -p .claude/skills
cp -r /tmp/mcp-gateway-registry/.claude/skills/agentcore-register .claude/skills/agentcore-register
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Agentcore Register in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Agentcore Register on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Agentcore Register is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

AgentCore Register

Given a remote MCP server URL, probe it to discover server info and tools, then generate a markdown file containing copy-pasteable values for every field in the Amazon Bedrock AgentCore Create record form.

The MCP server URL is provided as $ARGUMENTS.

If $ARGUMENTS is empty, ask the user for the MCP server URL using AskUserQuestion.

Steps

1. Initialize the MCP session

Send the MCP initialize request and capture the response headers (for the session ID) and body (for server info).

bash
curl -s --max-time 15 -D /tmp/_agentcore_mcp_headers.txt \
  -X POST "$MCP_URL" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "initialize",
    "params": {
      "protocolVersion": "2024-11-05",
      "capabilities": {},
      "clientInfo": {"name": "agentcore-register", "version": "1.0.0"}
    }
  }' > /tmp/_agentcore_mcp_init.json 2>&1

Parse the response to extract:

  • serverInfo.name - the server name
  • serverInfo.version - the server version
  • protocolVersion - the MCP protocol version
  • capabilities - what the server supports

Extract the mcp-session-id header value from /tmp/_agentcore_mcp_headers.txt.

Error handling
  • If the curl request fails or times out, tell the user the server is unreachable and stop.
  • If the response is a 502/503/504, tell the user the server backend is down and stop.
  • If the response is HTML (not JSON), tell the user the URL may not be an MCP endpoint and stop.

2. List the tools

Using the session ID from step 1, send a tools/list request:

bash
SESSION_ID=$(grep -i 'mcp-session-id' /tmp/_agentcore_mcp_headers.txt | awk '{print $2}' | tr -d '\r')

curl -s --max-time 15 \
  -X POST "$MCP_URL" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "mcp-session-id: $SESSION_ID" \
  -d '{"jsonrpc": "2.0", "id": 3, "method": "tools/list", "params": {}}' \
  > /tmp/_agentcore_mcp_tools.json 2>&1

The response may be plain JSON or SSE-formatted (data: {...}). Handle both:

  • If the response starts with {, parse it directly as JSON.
  • If the response contains data: lines, extract the JSON from the data: line.

Parse the tools array from result.tools. For each tool, capture:

  • name
  • description
  • inputSchema (the full JSON schema object)

3. Generate the markdown file

Create a markdown file at .scratchpad/agentcore-register-<server-name>.md with all the values the user needs to copy-paste into the AgentCore form.

Use this exact template, replacing placeholders with the discovered values:

markdown
# AgentCore Registry Record: <server_name>

*Generated: <current date>*
*Source: <MCP_URL>*

---

## Record Details

**Name:**

<server_name as a valid record ID, e.g. record_novacolor_finishes>


**Description:**

<server description derived from tools and server name, 1-2 sentences>


**Record version:**

<server version, e.g. 0.1>


**Record type:** MCP

---

## MCP Server Definition

Copy-paste the JSON below into the "Your MCP server definition" editor:

```json
{
  "name": "<namespace/slug, must match ^[a-zA-Z0-9.-]+/[a-zA-Z0-9._-]+$>",
  "description": "<max 100 chars, concise description>",
  "version": "<version>",
  "remotes": [
    {
      "type": "<transport type, e.g. streamable-http>",
      "url": "<MCP_URL>"
    }
  ]
}

Tool Definition

Copy-paste the JSON below into the "Your Tool definition" editor:

json
{
  "tools": [
    <for each tool, include the full tool object with name, description, and inputSchema>
  ]
}

Discovery Summary

FieldValue
Server name<serverInfo.name>
Version<version>
Protocol<protocolVersion>
Transport<transport type>
URL<MCP_URL>
Tools
Capabilities

#### Rules for populating the template

- **Record name**: Convert the server name to a valid AgentCore record ID. Use lowercase, replace spaces and hyphens with underscores, prefix with `record_`. Only use letters, digits, underscores. Example: `record_novacolor_finishes`.
- **Description**: Write a concise 1-2 sentence description based on the server name and tool names/descriptions.
- **Transport**: Determine from how the server responded:
  - If the initialize response included `mcp-session-id` header and responded to POST with JSON, use `streamable-http`.
  - If the server used SSE (event-stream responses for initialize), use `sse`.
  - Default to `streamable-http` for remote HTTP servers.
- **MCP server definition JSON**: Must include `name`, `description`, `version`, and `remotes` fields. The `remotes` array contains the transport type and URL so that the AgentCore sync can extract `proxy_pass_url` correctly via `remotes[0].url`.
  - **name** MUST match the pattern `^[a-zA-Z0-9.-]+/[a-zA-Z0-9._-]+$` (namespace/name format). Derive the namespace from the server's domain (e.g., `com.agenthost` for `agenthost.club`, `io.github.user` for GitHub). Example: `com.agenthost/novacolor-italian-finishes`.
  - **description** MUST be 100 characters or fewer. Keep it concise.
  - **remotes** MUST be an array with at least one object containing `type` (transport type) and `url` (the MCP endpoint URL). Do NOT use top-level `url` and `transport` fields -- use the `remotes` array format instead.
- **Tool definition JSON**: Include the complete tool objects exactly as returned by the server, wrapped in a `{"tools": [...]}` envelope.
- **Tool inputSchema**: Include the full inputSchema for each tool exactly as returned by the server. Do NOT simplify or omit properties.

### 4. Cleanup

Remove temporary files:

```bash
rm -f /tmp/_agentcore_mcp_headers.txt /tmp/_agentcore_mcp_init.json /tmp/_agentcore_mcp_tools.json

5. Report to the user

After generating the file, display:

  1. The path to the generated markdown file.
  2. A brief summary: server name, number of tools, transport type.
  3. Tell the user they can open the file and copy-paste each section into the AgentCore "Create record" form.

Example

User: /agentcore-register https://novacoloritalianfinishes.agenthost.club/mcp

Output:
Generated `.scratchpad/agentcore-register-novacolor.md`

Summary:
- Server: some-agent (v0.1)
- Transport: streamable-http
- Tools: 3 (search_site_products, search, fetch)

Open the file and copy-paste each section into the AgentCore "Create record" form.

Frequently asked questions

What does the Agentcore Register AI skill do?

Given an MCP server URL, probe the server via curl to discover its metadata and tools, then generate a markdown file with copy-pasteable content for each field in the Amazon Bedrock AgentCore "Create record" form.

Why use Agentcore Register on TypingMind?

Because you install it once and use it with any model. Agentcore Register is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Agentcore Register in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/agentic-community/mcp-gateway-registry/tree/main/.claude/skills/agentcore-register. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Agentcore Register?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Agentcore Register?

As many as you like. As long as a model supports skills, you can use Agentcore Register with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Agentcore Register AI skill free?

Yes. It is published on GitHub by agentic-community under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇