diff --git a/hive-agent-mcp/src/mcp/mod.rs b/hive-agent-mcp/src/mcp/mod.rs
index b233e03b..75a86481 100644
--- a/hive-agent-mcp/src/mcp/mod.rs
+++ b/hive-agent-mcp/src/mcp/mod.rs
@@ -347,15 +347,9 @@ impl AgentServer {
         at turn start to remember what you owe / what's owed to you without scrolling \
         inbox history. Output is a short bulleted list with ids, ages in seconds, and \
         the relevant context. Each `question` or `reminder` row can be cancelled by \
-        passing its id + kind to `cancel_loose_end`. Empty result is reported clearly.\n\
-        Todos are a to-do list, not an inbox to empty on sight — a row that's still \
-        genuinely in flight (a running bash task, say) is expected to sit there and \
-        doesn't need marking done early; only mark done the ones you've actually \
-        reviewed and resolved. A real content change (like a task finishing) resurfaces \
-        the same row on its own regardless of whether you marked it done while it was \
-        still running, so doing that buys nothing and just costs a redundant call later. \
-        Todos cap at 40 rows; a trailer line says how many more are pending — mark the \
-        reviewed ones done with `mark_todos_done`, then call again for the rest.\n\
+        passing its id + kind to `cancel_loose_end`. Empty result is reported clearly. Todos \
+        cap at 40 rows; a trailer line says how many more are pending — clear the shown ones \
+        with `mark_todos_done`, then call again for the rest.\n\
         Pass `agent: \"<name>\"` to inspect a specific peer agent's threads. Direct \
         child agents are always accessible. For non-children, the `query_agent_state` \
         capability is required — without it the request is rejected with an error."
@@ -555,19 +549,19 @@ impl AgentServer {
     }
 
     #[tool(
-        description = "Bulk-mark specific todos done by id in one call, instead of \
+        description = "Bulk-clear specific todos by id in one call, instead of \
         `cancel_loose_end`ing each one. List-based, not range-based — no `ack_until`-style \
-        'clear below id N' for todos, since a blind range-clear risks marking something \
-        done that you never actually reviewed. Pass the ids you've actually reviewed \
-        (typically what `get_loose_ends` just showed); unknown/already-done ids are \
-        silently skipped. Returns how many were newly marked done."
+        'clear below id N' for todos, since a blind range-clear risks acking something you \
+        never saw. Pass the ids you've actually reviewed (typically what `get_loose_ends` \
+        just showed); unknown/already-acked ids are silently skipped. Returns how many were \
+        newly acked."
     )]
     async fn mark_todos_done(&self, Parameters(args): Parameters<MarkTodosDoneArgs>) -> String {
         let log = format!("{args:?}");
         run_tool_envelope("mark_todos_done", log, async move {
             match mark_local_todos_done(args.ids).await {
                 Some(hive_agent_sock::Response::Acked { count }) => {
-                    format!("marked {count} todo(s) done")
+                    format!("acked {count} todo(s)")
                 }
                 Some(hive_agent_sock::Response::Err { message }) => {
                     format!("mark_todos_done failed: {message}")