Search a directory of 70K+ AI building blocks: MCP servers, agents, models, datasets, prompts.
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.
Full details of one listing as Markdown: description, type, source, license, popularity, links, similar listings and a comparison table.
Listing ID from search_listings, e.g. 'gh-langchain-ai-langchain-3b1c89'.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http stackmark https://stackmark.dev/mcp
Full details of one listing as Markdown: description, type, source, license, popularity, links, similar listings and a comparison table.
| Argument | Type | Description |
|---|---|---|
| id* | string | Listing ID from search_listings, e.g. 'gh-langchain-ai-langchain-3b1c89'. |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"id"
],
"properties": {
"id": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Listing ID from search_listings, e.g. 'gh-langchain-ai-langchain-3b1c89'."
}
},
"additionalProperties": false
}The types of building blocks on Stackmark, with how many public listings each has.
No arguments.
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {}
}Search AI building blocks on Stackmark by keyword, optionally limited to one type. Returns the most relevant listings first, with IDs for get_listing.
| Argument | Type | Description |
|---|---|---|
| type | string | Only this type of building block. |
| limit | integer | How many results (default 10). |
| query | string | Keywords, e.g. 'postgres', 'speech recognition'. Leave empty to list the most popular. |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {
"type": {
"enum": [
"prompt",
"skill",
"workflow",
"agent",
"tool",
"mcp_server",
"model",
"dataset",
"api",
"framework"
],
"type": "string",
"description": "Only this type of building block."
},
"limit": {
"type": "integer",
"maximum": 20,
"minimum": 1,
"description": "How many results (default 10)."
},
"query": {
"type": "string",
"maxLength": 160,
"description": "Keywords, e.g. 'postgres', 'speech recognition'. Leave empty to list the most popular."
}
},
"additionalProperties": false
}Up to 5 listings of the same type that are most similar to the given one; useful for finding alternatives.
| Argument | Type | Description |
|---|---|---|
| id* | string | Listing ID from search_listings. |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"id"
],
"properties": {
"id": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Listing ID from search_listings."
}
},
"additionalProperties": false
}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 13, 2026.