diff --git a/README.md b/README.md
index bfbf303..f9e1606 100644
--- a/README.md
+++ b/README.md
@@ -1,8 +1,7 @@
 # delegate-ai-mcp
 
-An MCP server that lets Claude Code delegate light, low-stakes work to a local
-[qwen-code](https://github.com/QwenLM/qwen-code) CLI instance, instead of spending its own
-budget on it.
+MCP servers that let Claude Code delegate light, low-stakes work to local AI CLI backends,
+instead of spending its own budget on it. Currently one backend: qwen-code.
 
 ## How it works
 
@@ -17,10 +16,18 @@ was confirmed. Replies can take several minutes on a cold local model, so the to
 generous (10 min) timeout rather than trying to enforce "light work" in code — that's a
 judgment call left to whoever's calling the tool.
 
+Each backend is its own MCP server process with its own `claude mcp add` registration —
+there's no single multi-tool server. Adding a second backend means adding its own
+`src/<backend>-delegate.ts` + `src/<backend>-delegate-server.ts` pair and registering it
+separately (see Status below — the shared subprocess helper this implies isn't extracted yet).
+
+No build step: everything runs as plain `.ts` source via Node's native TypeScript
+type-stripping. No `tsconfig.json`, no `dist/`, no lint/format tooling — deliberate choices,
+not oversights (see the wayfinder map linked below).
+
 ## Prerequisites
 
-- Node.js 24+ (the server runs `.ts` source directly via Node's native TypeScript
-  type-stripping — no build step, no compiler)
+- Node.js 24+ (native TypeScript type-stripping)
 - `qwen` on `PATH` and working (`qwen -p "test"` should return a reply)
 
 ## Setup
@@ -61,9 +68,12 @@ Runs `test/qwen-delegate.test.ts` on Node's built-in test runner (`node --test`)
 framework dependency. The subprocess spawn is mocked, so tests run in milliseconds and don't
 require a live `qwen` install.
 
-## Status / roadmap
+## Status
 
-Project conventions (build/typecheck strategy, lint/format tooling, multi-tool source
-structure, publish-shaped `package.json`) are being worked through a
+Project conventions are being worked through a
 [wayfinder map](https://git.arthurerlich.de/haylan/delegate-ai-mcp/issues/1) on this repo's
-issue tracker — see `docs/agents/issue-tracker.md` for how issues/tickets work here.
+issue tracker (see `docs/agents/issue-tracker.md` for how issues/tickets work here). Decided
+so far: no build process, ever; separate MCP server process + registration per backend, with a
+shared `src/lib/spawn-cli.ts` helper once a second backend exists (not extracted yet — still
+just `qwen-delegate.ts`); no lint/format tooling for now. Still open: a publish-shaped
+`package.json`.
diff --git a/src/qwen-delegate-server.ts b/src/qwen-delegate-server.ts
index c39145d..7162484 100644
--- a/src/qwen-delegate-server.ts
+++ b/src/qwen-delegate-server.ts
@@ -16,8 +16,7 @@ server.registerTool(
     title: "Delegate to qwen",
     description:
       "Delegate a light, low-stakes prompt (simple lookups, quick research, boilerplate text, " +
-      "small single-file edits) to the local qwen-code CLI, offloading it from the main session. " +
-      "Can take several minutes to respond — only use for work that doesn't block on a fast reply.",
+      "small single-file edits) to the local qwen-code CLI, offloading it from the main session. ",
     inputSchema: {
       prompt: z.string().describe("The prompt to send to qwen, non-interactively."),
     },
diff --git a/src/qwen-delegate.ts b/src/qwen-delegate.ts
index 2f8bfed..04317a9 100644
--- a/src/qwen-delegate.ts
+++ b/src/qwen-delegate.ts
@@ -10,7 +10,8 @@ export interface DelegateResult {
 }
 
 export interface DelegateOptions {
-  /** Milliseconds before giving up. qwen-code has been observed taking 3-6 min for a trivial
+  /** Milliseconds before giving up. qwen-code has been observed taking 3-6 min for a trivial, 
+   * takes time to warm up on the first request,
    * prompt against a local model proxy, so default generous — see research/qwen-mcp-delegation.md. */
   timeoutMs?: number;
   /** Injectable for tests. Defaults to node:child_process's spawn. */
@@ -24,7 +25,7 @@ const DEFAULT_TIMEOUT_MS = 10 * 60 * 1000; // 10 min
  * qwen-code prints startup warnings (e.g. failed MCP sub-servers) to stderr; those are
  * ignored on success and surfaced only when the run itself fails.
  */
-export function delegateToQwen(
+export function delegateToQwen (
   prompt: string,
   options: DelegateOptions = {},
 ): Promise<DelegateResult> {