ai.promptarch/mcp — Model Context Protocol (MCP) Server

The ai.promptarch/mcp server provides research-backed linting and generation for agent context files, targeting formats such as CLAUDE.md, AGENTS.md, and Cursor rules. It is exposed via an MCP interface and includes tools to support these workflows.

🛠️ Key Features

Research-backed linting for agent context files
Generation support for agent context files
Targets CLAUDE.md, AGENTS.md, and Cursor rules
Tooling available via 3 MCP tools

🚀 Use Cases

Validate and lint CLAUDE.md agent context files
Create or generate AGENTS.md content
Maintain or generate Cursor rules for agent behavior

⚡ Developer Benefits

Consistent formatting for agent context files
Support for common agent-context document types

⚠️ Limitations

Scope described only for linting and generation of agent context files (CLAUDE.md, AGENTS.md, Cursor rules)
No additional tools or integrations beyond the 3 listed tools are specified

{ "type": "object", "properties": { "content": { "type": "string", "description": "The full file content to lint." }, "format": { "description": "Override format auto-detection.", "type": "string", "enum": [ "claude_md", "agents_md", "cursor_mdc", "copilot_instructions", "memory_file" ] }, "filename": { "description": "Filename hint for format detection (e.g. CLAUDE.md).", "type": "string" } }, "required": [ "content" ], "$schema": "http://json-schema.org/draft-07/schema#" }

{ "type": "object", "properties": { "artifact": { "type": "string", "description": "Artifact id. See promptarch_list_artifact_types (e.g. context_pack)." }, "project_name": { "type": "string" }, "project_description": { "type": "string" }, "tech_stack": { "type": "string" }, "repo_structure": { "type": "string" }, "commands": { "type": "string" }, "coding_conventions": { "type": "string" }, "target_model": { "description": "Target model family (default Claude).", "type": "string" } }, "required": [ "artifact", "project_name", "project_description" ], "$schema": "http://json-schema.org/draft-07/schema#" }

Install

Tools (3)

promptarch_lint_artifactpromptarch_list_artifact_typespromptarch_generate_artifact

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://mcp.promptarch.ai/mcp

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.promptarch.ai/mcp"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

content	string	required	The full file content to lint.
format	string	optional	Override format auto-detection.
filename	string	optional	Filename hint for format detection (e.g. CLAUDE.md).

artifact	string	required	Artifact id. See promptarch_list_artifact_types (e.g. context_pack).
project_name	string	required
project_description	string	required
tech_stack	string	optional
repo_structure	string	optional
commands	string	optional
coding_conventions	string	optional
target_model	string	optional	Target model family (default Claude).

ai.promptarch/mcp — Model Context Protocol (MCP) Server

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

promptarch_lint_artifact

promptarch_list_artifact_types

promptarch_generate_artifact

Install

Remote endpoint

Related servers