fix(agent): fully silent assent — no system notes, no chat_assent events

The len(pending)>0 path still injected a brief note saying 'execution(s)
are now running' — the model saw this, thought work was being done for
it, and no-op'd (finish_reason=stop, content_len=0). Same confusion as
the len(pending)==0 case, just from the other branch.

Fix: both assent paths are now fully silent. No system note at all. The
model sees 'go ahead' in the replayed history and responds naturally.

Also removed chat_assent tool_use/tool_result emit events. These were
persisted in the transcript and confused the model on replay — it saw
its own 'tool calls' (chat_assent) and thought it had already acted.

VERSION 0.7.4 → 0.7.5

VERSION

@@ -1 +1 @@
-0.7.4
+0.7.5

cmd/nomos/agent.go

@@ -296,17 +296,6 @@ func (a *agent) chatWith(ctx context.Context, sessionID, message, systemInject s
 			if ok {
 				granted = append(granted, p.execID)
 				slog.Info("nomos: chat-assent granted", "execution", p.execID, "status", status, "session", sessionID)
-				// Mark as continued so the continuation worker doesn't
-				// pick up this execution and call resumeSession while the
-				// chat handler is still processing "go ahead." Without this,
-				// two LLM calls run concurrently for the same session —
-				// the chat handler's chat() and the worker's resumeSession()
-				// — causing empty responses and race conditions.
-				if execUUID, perr := uuid.Parse(p.execID); perr == nil {
-					a.store.markContinued(ctx, execUUID)
-				}
-				emit(agentEvent{Type: "tool_use", Data: map[string]any{"name": "chat_assent", "args": map[string]any{"execution_id": p.execID}, "id": "assent-" + p.execID}, SessionID: sessionID})
-				emit(agentEvent{Type: "tool_result", Data: map[string]any{"name": "chat_assent", "result": fmt.Sprintf("Approved via chat assent (%q). Status: %s.", message, status), "id": "assent-" + p.execID}, SessionID: sessionID})
 
 				// An explicit typed confirmation for a destructive action
 				// opens a short, target-scoped window so the rest of a
@@ -324,15 +313,22 @@ func (a *agent) chatWith(ctx context.Context, sessionID, message, systemInject s
 		}
 		if len(granted) > 0 {
 			a.openAssentWindow(ctx, sessionID)
-			// Brief note: only tells the model WHICH specific executions
-			// were approved (so it doesn't re-request them). Does NOT say
-			// "continue the plan" — the model already sees "go ahead" in
-			// the replayed history and knows to continue. The old verbose
-			// note ("they are now running... you MUST continue...") made
-			// the model think work was being done for it, causing empty
-			// responses (finish_reason=stop, content_len=0).
-			note := fmt.Sprintf("[System: execution(s) %s were approved by the operator and are now running. Do not re-request those — check get_execution_status if you need the outcome.]", strings.Join(granted, ", "))
-			messages = append(messages, openai.SystemMessage(note))
+			// Mark approved executions as continued so the continuation
+			// worker doesn't call resumeSession while the chat handler is
+			// still processing "go ahead" — two concurrent LLM calls for the
+			// same session cause empty responses and race conditions.
+			for _, execID := range granted {
+				if execUUID, perr := uuid.Parse(execID); perr == nil {
+					a.store.markContinued(ctx, execUUID)
+				}
+			}
+			// No system note. The model already sees "go ahead" in the
+			// replayed history (the user message was saved to the DB before
+			// chat() was called). The old note said "they are now running"
+			// which made the model think work was being done for it —
+			// causing empty responses (finish_reason=stop, content_len=0).
+			// The approved executions are dispatched; the model will
+			// continue with the remaining plan steps naturally.
 		}
 		if len(blocked) > 0 {
 			note := fmt.Sprintf("[System: execution(s) %s are classified DESTRUCTIVE and were NOT approved by loose assent — you must ask the operator for an explicit typed confirmation before they can run. Once they do confirm, further destructive steps on that SAME target (e.g. finishing a stop-then-destroy sequence) will auto-run for 15 minutes without asking again — but a different target always needs its own confirmation.]", strings.Join(blocked, ", "))
@@ -341,15 +337,9 @@ func (a *agent) chatWith(ctx context.Context, sessionID, message, systemInject s
 	} else if assent && len(pending) == 0 {
 		// The operator said "proceed"/"go ahead"/"yes" but there are no
 		// pending approvals — the agent proposed a plan (via propose_plan)
-		// and asked "shall I?" Open the assent window silently. Do NOT
-		// inject a system note: the model already sees "go ahead" in the
-		// replayed history (the user message was saved to the DB before
-		// chat() was called, and getRecentMessages replays it). The old
-		// verbose system note ("The operator approved your proposed plan.
-		// Execute it now — call update_plan_step then run...") was redundant
-		// with the user's "go ahead" and caused the model to no-op
-		// (finish_reason=stop, content_len=0) — the model saw "approved" +
-		// "running" and concluded there was nothing to do.
+		// and asked "shall I?" Open the assent window silently. No system
+		// note: the model sees "go ahead" in the replayed history and
+		// responds naturally.
 		a.openAssentWindow(ctx, sessionID)
 	}