Shinkuro

# Shinkuro - Universal prompt loader MCP server [](https://pypi.org/project/shinkuro/)  Loads markdown files from a local folder or git repository and serves them as [MCP Prompts](https://modelcontextprotocol.io/specification/2025-06-18/server/prompts). Useful for loading prompts from various sources and formats into your MCP-enabled applications, and sharing prompts across organizations. <!-- waiting the license issue to be fixed --> <!-- <a href="https://glama.ai/mcp/servers/@DiscreteTom/shinkuro"> <img width="380" height="200" src="https://glama.ai/mcp/servers/@DiscreteTom/shinkuro/badge" alt="Shinkuro MCP server" /> </a> --> ## Usage **IMPORTANT**: make sure your MCP client supports the MCP Prompts capability. See the [feature support matrix](https://modelcontextprotocol.io/clients#feature-support-matrix). ### Full CLI Usage <details> <summary><code>uvx shinkuro --help</code></summary> ```sh Usage: shinkuro [OPTIONS] Shinkuro - Universal prompt loader MCP server ╭─ Options ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮ │ --folder TEXT Path to local folder containing markdown files, or subfolder within git repo [env var: FOLDER] │ │ --git-url TEXT Git repository URL (supports GitHub, GitLab, SSH, HTTPS with credentials) [env var: GIT_URL] │ │ --cache-dir TEXT Directory to cache remote repositories [env var: CACHE_DIR] [default: ~/.shinkuro/remote] │ │ --auto-pull Whether to refresh local cache on startup [env var: AUTO_PULL] │ │ --variable-format [brace|dollar] Template variable format [env var: VARIABLE_FORMAT] [default: brace] │ │ --auto-discover-args Auto-discover template variables as required arguments [env var: AUTO_DISCOVER_ARGS] │ │ --skip-frontmatter Skip frontmatter processing and use raw markdown content [env var: SKIP_FRONTMATTER] │ │ --version Show version and exit │ │ --help Show this message and exit. │ ╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯ ``` </details> ### Local Files Add to your MCP client configuration: ```json { "mcpServers": { "shinkuro": { "command": "uvx", "args": ["shinkuro"], "env": { "FOLDER": "/path/to/prompts" } } } } ``` ### Remote Git Repository Add to your MCP client configuration: ```json { "mcpServers": { "shinkuro": { "command": "uvx", "args": ["shinkuro"], "env": { "GIT_URL": "https://github.com/owner/repo.git", "FOLDER": "prompts" // optional, subfolder within git repo } } } } ``` > This will clone the repository into a local cache dir. Make sure you have correct permission. > Private repositories are supported, e.g. `"GIT_URL": "git@github.com:DiscreteTom/shinkuro.git"` (with SSH keys), `"GIT_URL": "https://<username>:<PAT>@github.com/owner/repo.git"` (with personal access token) ### Use with [Spec-Kit](https://github.com/github/spec-kit) <details> <summary>Expand</summary> First, move spec-kit prompts into `./.shinkuro/prompts` folder. Then add to your MCP client configuration: ```json { "mcpServers": { "shinkuro": { "command": "uvx", "args": ["shinkuro"], "env": { "FOLDER": "./.shinkuro/prompts", "VARIABLE_FORMAT": "dollar", "AUTO_DISCOVER_ARGS": "true", "SKIP_FRONTMATTER": "true" } } } } ``` This will expose spec-kit instructions as MCP prompts. </details> ## Prompt Loading Each markdown file in the specified folder (including nested folders) is loaded as a prompt. Example folder structure: ``` my-prompts/ ├── think.md └── dev/ ├── code-review.md └── commit.md ``` The example above will be loaded to 3 prompts: `think`, `code-review` and `commit`. ## Example Prompt Files ### Simplest ```markdown Commit to git using conventional commit. ``` ### Prompt with Metadata ```markdown --- name: "code-review" # optional, defaults to filename title: "Code Review Assistant" # optional, defaults to filename description: "" # optional, defaults to file path --- # Code Review Please review this code for best practices and potential issues. ``` ### Prompt with Arguments ```markdown --- name: "greeting" description: "Generate a personalized greeting message" arguments: - name: "user" description: "Name of the user" # no default = required parameter - name: "project" description: "Project name" default: "MyApp" --- Say: Hello {user}! Welcome to {project}. Hope you enjoy your stay! ``` Variables like `{user}` and `{project}` will be replaced with actual values when the prompt is retrieved. Use `{{var}}` (double brackets) to escape and display literal brackets when using brace formatter. > **Different Variable Formats:** > > - `brace` (default): `{user}`, `{project}` > - `dollar`: `$user`, `$project` ## Example Prompt Repositories - [DiscreteTom/prompts](https://github.com/DiscreteTom/prompts). ## [CHANGELOG](./CHANGELOG.md)