Struct SoloMcpServer 

pub struct SoloMcpServer { /* private fields */ }

The MCP server. Cheap to clone — every field is Arc-cloneable.

v0.8.0 P2: an MCP session resolves to one tenant. The session’s tenant_handle is resolved at initialize time (today: from the CLI invocation via solo mcp-stdio --tenant <id>; future versions may resolve per-bearer-token via OIDC). Subsequent tools/call invocations route through the cached handle without re-resolving. Operators that need multi-tenant MCP spawn one solo mcp-stdio subprocess per tenant.

Implementations

impl SoloMcpServer

pub fn new_for_tenant( registry: Arc<TenantRegistry>, tenant: Arc<TenantHandle>, user_aliases: Vec<String>, ) -> Self

Build a server speaking for tenant (v0.8.0 P2 — one MCP session ↔ one tenant). The registry is held so future capabilities can reach across tenants if needed; today every handler routes through self.inner.tenant.

v0.8.1 P2: auto-resolves the audit principal from the SOLO_MCP_PRINCIPAL_TOKEN env var (see resolve_mcp_principal). When neither the env var nor a header is set, the principal stays None — preserving v0.8.0 behavior for single-user setups.

pub fn new_for_tenant_with_principal( registry: Arc<TenantRegistry>, tenant: Arc<TenantHandle>, user_aliases: Vec<String>, audit_principal: Option<String>, ) -> Self

v0.8.0 P4: like Self::new_for_tenant, but records an explicit audit principal subject for every tool dispatch. MCP is bearer-only at v0.8.0 — the orchestration layer (today: the daemon’s --bearer-token-file path) decides whether a session counts as “bearer-authenticated” and passes Some("bearer"); CLI / unauth paths pass None.

v0.8.1 P2: when the caller passes audit_principal = None, the env-var auto-resolution still runs (in new_for_tenant). Callers who want to explicitly suppress env-var resolution can call this method with None after std::env::remove_var(...), or use the dedicated test constructor that bypasses env reads.

impl SoloMcpServer

pub async fn dispatch_tool( &self, name: &str, args_value: Value, progress: Option<ProgressReporter>, ) -> Result<CallToolResult, McpError>

Direct tool-dispatch path used by both call_tool (the ServerHandler trait method, behind the rmcp protocol layer) and in-process tests that don’t want to spin up a full transport pair. Bypasses RequestContext (which requires a Peer not constructible outside rmcp internals).

v0.11.0 P3: progress is Some only when the HTTP transport dispatched the request AND the client opted in via _meta.progressToken. The three long-running handlers (memory_ingest_document, memory_search_docs, memory_remember_batch) consult the reporter; the other 11 handlers ignore it (backward compat with stdio and with HTTP clients that did not opt in).

pub fn dispatch_list_tools(&self) -> Vec<Tool>

List the tools this server exposes. Mirrors ServerHandler::list_tools without requiring a RequestContext.

Trait Implementations

impl Clone for SoloMcpServer

fn clone(&self) -> SoloMcpServer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl ServerHandler for SoloMcpServer

async fn initialize( &self, request: InitializeRequestParams, context: RequestContext<RoleServer>, ) -> Result<InitializeResult, McpError>

v0.9.0 P2: override initialize so we can:

1. Cache the client’s InitializeRequestParams on the peer (delegates to rmcp’s default for this).
2. If the tenant’s [llm] mode = "mcp_sampling": a. Refuse to initialize when the peer didn’t advertise the sampling capability — surfaces the BLOCKER 2-locked error message so the user sees commented-out alternative TOML blocks. b. Otherwise build a SamplingLlmClient-backed Steward and write it into tenant.steward_slot() so the writer actor’s next consolidate-tick reads a populated slot.
3. For any other [llm] mode, return the configured tools surface unchanged (the slot was eagerly populated at registry-open time by the static StewardFactory).

fn get_info(&self) -> ServerInfo

async fn list_tools( &self, _request: Option<PaginatedRequestParam>, _context: RequestContext<RoleServer>, ) -> Result<ListToolsResult, McpError>

async fn call_tool( &self, request: CallToolRequestParam, _context: RequestContext<RoleServer>, ) -> Result<CallToolResult, McpError>

fn enqueue_task( &self, _request: CallToolRequestParams, _context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<CreateTaskResult, ErrorData>> + MaybeSendFuture

fn ping( &self, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<(), ErrorData>> + MaybeSendFuture

fn complete( &self, request: CompleteRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<CompleteResult, ErrorData>> + MaybeSendFuture

fn set_level( &self, request: SetLevelRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<(), ErrorData>> + MaybeSendFuture

fn get_prompt( &self, request: GetPromptRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<GetPromptResult, ErrorData>> + MaybeSendFuture

fn list_prompts( &self, request: Option<PaginatedRequestParams>, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<ListPromptsResult, ErrorData>> + MaybeSendFuture

fn list_resources( &self, request: Option<PaginatedRequestParams>, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<ListResourcesResult, ErrorData>> + MaybeSendFuture

fn list_resource_templates( &self, request: Option<PaginatedRequestParams>, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<ListResourceTemplatesResult, ErrorData>> + MaybeSendFuture

fn read_resource( &self, request: ReadResourceRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<ReadResourceResult, ErrorData>> + MaybeSendFuture

fn subscribe( &self, request: SubscribeRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<(), ErrorData>> + MaybeSendFuture

fn unsubscribe( &self, request: UnsubscribeRequestParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<(), ErrorData>> + MaybeSendFuture

fn get_tool(&self, _name: &str) -> Option<Tool>

Get a tool definition by name.

fn on_custom_request( &self, request: CustomRequest, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<CustomResult, ErrorData>> + MaybeSendFuture

fn on_cancelled( &self, notification: CancelledNotificationParam, context: NotificationContext<RoleServer>, ) -> impl Future<Output = ()> + MaybeSendFuture

fn on_progress( &self, notification: ProgressNotificationParam, context: NotificationContext<RoleServer>, ) -> impl Future<Output = ()> + MaybeSendFuture

fn on_initialized( &self, context: NotificationContext<RoleServer>, ) -> impl Future<Output = ()> + MaybeSendFuture

fn on_roots_list_changed( &self, context: NotificationContext<RoleServer>, ) -> impl Future<Output = ()> + MaybeSendFuture

fn on_custom_notification( &self, notification: CustomNotification, context: NotificationContext<RoleServer>, ) -> impl Future<Output = ()> + MaybeSendFuture

fn list_tasks( &self, request: Option<PaginatedRequestParams>, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<ListTasksResult, ErrorData>> + MaybeSendFuture

fn get_task_info( &self, request: GetTaskInfoParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<GetTaskResult, ErrorData>> + MaybeSendFuture

fn get_task_result( &self, request: GetTaskResultParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<GetTaskPayloadResult, ErrorData>> + MaybeSendFuture

fn cancel_task( &self, request: CancelTaskParams, context: RequestContext<RoleServer>, ) -> impl Future<Output = Result<CancelTaskResult, ErrorData>> + MaybeSendFuture