ruby-sdk/AGENTS.md
Koichi ITO 719e5e7fee Introduce ServerSession for per-connection state
## Motivation and Context

The Ruby SDK uses a 1-Server-to-many-sessions model where session-specific state
(`@client`, `@logging_message_notification`) is stored on the shared `Server` instance.
This causes the last client to connect to overwrite state for all sessions,
and requires `session_id` to be threaded through method signatures for session-scoped notifications.

The Python SDK avoids this by creating a `ServerSession` per connection that wraps a shared `Server`.
Each session holds its own state and naturally scopes notifications through its own stream.

### Changes

- Added `MCP::ServerSession` class holding per-session state: client info,
  logging configuration, and a `transport` reference for session-scoped
  notification delivery.
- `StreamableHTTPTransport#handle_initialization` creates a `ServerSession`
  per `initialize` request and stores it in the session hash.
- `handle_regular_request` routes requests through the session's
  `ServerSession` instead of the shared `Server` directly.
- `Server#handle` and `Server#handle_json` accept an optional `session:`
  keyword (`ServerSession` instance).
- `Server#init` stores client info on the session when available.
- `Server#configure_logging_level` stores logging config on the session.
- `ServerContext` and `Progress` accept a `notification_target:` which
  can be either a `ServerSession` (session-scoped) or a `Server` (broadcast).
  No `session_id:` parameter needed.
- `ServerContext#notify_progress` and `#notify_log_message` delegate to
  the `notification_target` without `session_id` threading.
- `StdioTransport` creates a single `ServerSession` on `open`,
  making the session model transparent across both transports.

### Design Decision

This follows the Python SDK's "shared `Server` + per-connection `Session`" pattern.
The `Server` holds configuration (tools, prompts, resources, handlers).
Each `ServerSession` holds per-connection state (client info, logging level, stream writer).
Notifications from `ServerSession` go only to that session's stream.

## How Has This Been Tested?

All existing tests pass. All conformance tests pass. Added tests verifying:

- Session-scoped log notification is sent only to the originating session.
- Session-scoped progress notification is sent only to the originating session.
- Each session stores its own client info independently.
- Each session stores its own logging level independently.
- `StdioTransport#open` creates a `ServerSession` and stores client info on it.

## Breaking Change

None for end users.

The public API (`Server.new`, `define_tool`,`server_context.report_progress`,
`server_context.notify_log_message`, etc.) is unchanged. The following internal API
changes affect only SDK internals:

- `ServerContext.new` now requires `notification_target:` instead of just`progress:`.
- `Progress.new` now takes `notification_target:` instead of `server:`.
- `Server#handle` and `Server#handle_json` accept an optional `session:` keyword.
2026-03-29 08:04:56 +09:00

121 lines
4.8 KiB
Markdown

# AGENTS.md
## Project overview
This is the official Ruby SDK for the Model Context Protocol (MCP), implementing both server and client functionality for JSON-RPC 2.0 based communication between LLM applications and context providers.
## Dev environment setup
- Ruby 3.2.0+ required to run the full test suite, including all Sorbet-related features
- Run `bundle install` to install dependencies
- Dependencies: `json-schema` >= 4.1 - Schema validation
## Build and test commands
- `bundle install` - Install dependencies
- `rake test` - Run all tests
- `rake rubocop` - Run linter
- `rake` - Run tests and linting (default task)
- `ruby -I lib -I test test/path/to/specific_test.rb` - Run single test file
- `gem build mcp.gemspec` - Build the gem
## Testing instructions
- Test files are in `test/` directory with `_test.rb` suffix
- Run full test suite with `rake test`
- Run individual tests with `ruby -I lib -I test test/path/to/file_test.rb`
- Tests should pass before submitting PRs
## Code style guidelines
- Follow RuboCop rules (run `rake rubocop`)
- Use frozen string literals
- Follow Ruby community conventions
- Keep dependencies minimal
## Commit message conventions
- Use conventional commit format when possible
- Include clear, descriptive commit messages
- Releases are triggered by updating version in `lib/mcp/version.rb` and merging to main
## Release process
- Follow [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) format in CHANGELOG.md
- Update CHANGELOG.md before cutting releases
- Use git history and PR merge commits to construct changelog entries
- Format entries as: "Terse description of the change (#nnn)"
- Keep entries in flat list format (no nesting)
- Git tags mark commits that cut new releases
- Exclude maintenance PRs that don't concern end users
- Check upstream remote for PRs if available
## Architecture overview
### Core Components
**MCP::Server** (`lib/mcp/server.rb`):
- Main server class handling JSON-RPC requests and holding shared configuration (tools, prompts, resources, handlers, capabilities)
- Implements MCP protocol methods: initialize, ping, tools/list, tools/call, prompts/list, prompts/get, resources/list, resources/read
- Supports custom method registration via `define_custom_method`
- Handles instrumentation, exception reporting, and notifications
- Uses JsonRpcHandler for request processing
**MCP::ServerSession** (`lib/mcp/server_session.rb`):
- Per-connection state: client info, logging level
- Created by the transport layer for each client connection
- Delegates request handling to the shared `Server`
**MCP::Client** (`lib/mcp/client.rb`):
- Client interface for communicating with MCP servers
- Transport-agnostic design with pluggable transport layers
- Supports tool listing and invocation
**Transport Layer**:
- `MCP::Server::Transports::StdioTransport` - Command-line stdio transport
- `MCP::Server::Transports::StreamableHttpTransport` - HTTP with streaming support
- `MCP::Client::HTTP` - HTTP client transport (requires faraday gem)
**Protocol Components**:
- `MCP::Tool` - Tool definition with input/output schemas and annotations
- `MCP::Prompt` - Prompt templates with argument validation
- `MCP::Resource` - Resource registration and retrieval
- `MCP::Configuration` - Global configuration with exception reporting and instrumentation
### Key Patterns
**Three Ways to Define Components**:
1. Class inheritance (e.g., `class MyTool < MCP::Tool`)
2. Define methods (e.g., `MCP::Tool.define(name: "my_tool") { ... }`)
3. Server registration (e.g., `server.define_tool(name: "my_tool") { ... }`)
**Schema Validation**:
- Tools support input_schema and output_schema for JSON Schema validation
- Protocol version 2025-03-26+ supports tool annotations (destructive_hint, idempotent_hint, etc.)
- Validation is configurable via `configuration.validate_tool_call_arguments`
**Session Architecture**:
- `Server` holds shared configuration (tools, prompts, resources, handlers)
- `ServerSession` holds per-connection state (client info, logging level)
- Both `StdioTransport` and `StreamableHTTPTransport` create a `ServerSession` per connection, making the session model transparent across transports
- Session-scoped notifications (`notify_progress`, `notify_log_message`) are sent only to the originating client via `ServerSession`
- Server-wide notifications (`notify_tools_list_changed`, etc.) broadcast to all sessions via `Server`
**Context Passing**:
- `server_context` hash passed through tool/prompt calls for request-specific data
- Methods can accept `server_context:` keyword argument for accessing context
### Integration patterns
- **Rails controllers**: Use `server.handle_json(request.body.read)` for HTTP endpoints
- **Command-line tools**: Use `StdioTransport.new(server).open` for CLI applications
- **HTTP services**: Use `StreamableHttpTransport` for web-based servers