mirror of
https://github.com/modelcontextprotocol/servers.git
synced 2026-04-18 16:13:22 +02:00
Add instructions
This commit is contained in:
39
src/everything/instructions.md
Normal file
39
src/everything/instructions.md
Normal file
@@ -0,0 +1,39 @@
|
||||
# Everything Server
|
||||
|
||||
This is a comprehensive MCP server that demonstrates all major MCP features and capabilities.
|
||||
|
||||
## Available Tools
|
||||
|
||||
- **echo**: Echo back any message
|
||||
- **add**: Add two numbers together
|
||||
- **longRunningOperation**: Demonstrate progress notifications with configurable duration and steps
|
||||
- **printEnv**: Display all environment variables for debugging server configuration
|
||||
- **sampleLLM**: Request LLM sampling from the client with a custom prompt
|
||||
- **getTinyImage**: Return a small example image in PNG format
|
||||
- **annotatedMessage**: Show how content annotations work for different message types
|
||||
- **getResourceReference**: Return an embedded resource reference by ID (1-100)
|
||||
|
||||
## Available Resources
|
||||
|
||||
- Static resources numbered 1-100 accessible via `test://static/resource/{id}`
|
||||
- Even-numbered resources contain plain text, odd-numbered contain binary data
|
||||
- Resources support subscription for real-time updates every 10 seconds
|
||||
- Resource templates allow dynamic URI construction
|
||||
|
||||
## Available Prompts
|
||||
|
||||
- **simple_prompt**: Basic prompt without arguments
|
||||
- **complex_prompt**: Advanced prompt with temperature and style arguments, includes image content
|
||||
- **resource_prompt**: Prompt that embeds a specific resource by ID
|
||||
|
||||
## Features
|
||||
|
||||
- Pagination support for large resource lists
|
||||
- Argument completion for prompt parameters and resource IDs
|
||||
- Progress notifications for long-running operations
|
||||
- Configurable logging levels with automatic log message generation
|
||||
- Resource subscription system with periodic update notifications
|
||||
- LLM sampling integration for server-initiated model requests
|
||||
- Content annotations for priority and audience targeting
|
||||
|
||||
Use this server to test MCP client implementations and explore all protocol features.
|
||||
Reference in New Issue
Block a user