an act with nothing on the other end says so (V-556)

askClarify parked "Что сделать?" whatever was on the other end. With an empty
allowlist that question has no answer: she asks, fails, asks again and gives up,
three turns spent on a request she could have declined in the first one.

Empty allowlist now names the gap and parks nothing. A non-empty one still asks,
and names what she can run, capped at six, so the question is answerable.

cmd/mavend/clarifytemplates.go

@@ -1,6 +1,8 @@
 package main
 
 import (
+	"strings"
+
 	"github.com/kami/maven/internal/dialogue"
 	"github.com/kami/maven/internal/router"
 )
@@ -47,6 +49,36 @@ var clarifyQuestionVariants = map[dialogue.Slot][]string{
 	},
 }
 
+// namedActsCap bounds how many capability names one question may recite. Six is
+// what a spoken sentence carries; past that the list stops being an answer and
+// becomes a wall he has to hold in his head.
+const namedActsCap = 6
+
+// fnClarify decides what to say about an act whose capability is missing, and
+// whether the request is worth parking (Vikunja #556).
+//
+// The old deck asked "Что сделать?" whatever was on the other end. With an empty
+// allowlist that question has no answer: nothing he says can match, so she asks,
+// fails, asks again and gives up — three turns spent on a request she could have
+// declined in the first one. So an empty allowlist names the gap and parks
+// nothing, and a non-empty one asks a question he can actually answer by naming
+// what she has.
+func fnClarify(allow []string, attempt int) (reply string, ask bool) {
+	if len(allow) == 0 {
+		return "Я пока ничего не умею делать — мне не разрешён ни один инструмент.", false
+	}
+	q, ok := clarifyQuestionFor(dialogue.SlotFn, attempt)
+	if !ok {
+		return "", false
+	}
+	named := allow
+	tail := ""
+	if len(named) > namedActsCap {
+		named, tail = named[:namedActsCap], "…"
+	}
+	return q + " Я умею: " + strings.Join(named, ", ") + tail + ".", true
+}
+
 // clarifyQuestionFor picks the wording for this attempt. attempt is 1-based, as
 // PendingQuestion.Attempts counts it; anything past the list uses the last and
 // most explicit phrasing rather than wrapping round to the short one, because