Add ask_expert and eval capture (A14/A16) — step 7c

Completes step 7.

- expert.py: one stateless call for a question needing no repo access.
  Structured output is contracted at the prompt level, since vendors
  differ on JSON enforcement, so the reply is parsed and validated
  locally with one retry — tolerating a fenced code block, because
  models wrap JSON habitually, but not missing fields
- The per-task budget is counted from the canonical log rather than
  memory, so it survives a restart and cannot hand a circling executor
  a fresh allowance. Exhaustion raises rather than answering: a caller
  out of consults should get on with the work or escalate
- Consults are recorded as delegation_started/completed with role
  consultant. A16 asks for a "consult/delegation eval set", and
  reusing the composite events gets consult evaluation for free rather
  than inventing a parallel record
- evals.py pairs starts with completions into eval records, with
  verdicts kept outside the canonical log: the log records what
  happened, a verdict is a later judgement about it, and judgements
  get revised. Schema v3 adds the outcomes table, so a v1 store now
  traverses two migrations — a better exercise of the hook than one
- ask_expert is offered to the executor as a tool only when a desk is
  configured, in addition to the observable escalation triggers and
  never instead of them (A12). Consulting is not mutating, so a
  completion claim backed only by consults is still challenged

Found live: passing an empty message list and putting the question in
the system prompt fails outright — the Messages API requires at least
one non-system message. The question now travels as a synthesised user
message that is never appended to the log, which keeps the consult
stateless while remaining a valid request.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

nisse-dev/tests/test_executor.py

@@ -223,3 +223,95 @@ def test_multiple_calls_in_one_turn_all_run(store: StateStore, ws: Workspace) ->
         e.event for e in store.read_events() if type(e.event).__name__ == "ToolResult"
     ]
     assert {r.call_id for r in results} == {"a", "b"}
+
+
+class StubDesk:
+    """Stands in for ExpertDesk without a model behind it."""
+
+    def __init__(self, answer: str = "use sqlite", error: Exception | None = None):
+        self.asked: list[str] = []
+        self._answer = answer
+        self._error = error
+
+    def ask(self, question: str, **kw: Any) -> Any:
+        self.asked.append(question)
+        if self._error is not None:
+            raise self._error
+
+        class _A:
+            def render(self_inner) -> str:
+                return self._answer
+
+        return _A()
+
+
+def test_ask_expert_is_offered_only_when_a_desk_exists(
+    store: StateStore, ws: Workspace
+) -> None:
+    client = ScriptedClient([])
+    build(client, store, ws).run(milestone=1)
+    assert all(t["function"]["name"] != "ask_expert" for t in client.calls[0]["tools"])
+
+    client2 = ScriptedClient([])
+    build(client2, store, ws, expert=StubDesk()).run(milestone=1)
+    names = [t["function"]["name"] for t in client2.calls[0]["tools"]]
+    assert names[-1] == "ask_expert"
+    # A12: in addition to the observable triggers, not instead of them.
+    assert "run_command" in names
+
+
+def test_ask_expert_answer_comes_back_as_a_tool_result(
+    store: StateStore, ws: Workspace
+) -> None:
+    desk = StubDesk(answer="prefer sqlite here")
+    client = ScriptedClient(
+        [
+            [call("ask_expert", question="sqlite or postgres?")],
+            [call("write_file", path="a.py", content="x")],
+            [AssistantText(text="done", model="m")],
+        ]
+    )
+    build(client, store, ws, expert=desk).run(milestone=1)
+
+    assert desk.asked == ["sqlite or postgres?"]
+    results = [
+        e.event for e in store.read_events() if type(e.event).__name__ == "ToolResult"
+    ]
+    assert "prefer sqlite here" in results[0].result
+
+
+def test_consulting_is_not_doing_the_work(store: StateStore, ws: Workspace) -> None:
+    # Asking is not changing anything, so a completion claim backed only by
+    # consults is still challenged.
+    client = ScriptedClient(
+        [
+            [call("ask_expert", question="what should I do?")],
+            [AssistantText(text="I think it is done", model="m")],
+            [AssistantText(text="really done", model="m")],
+        ]
+    )
+    outcome = build(client, store, ws, expert=StubDesk()).run(milestone=1)
+    assert outcome.status == "escalate"
+
+
+def test_exhausted_budget_is_reported_to_the_model_not_raised(
+    store: StateStore, ws: Workspace
+) -> None:
+    from nisse_dev.expert import BudgetExhausted
+
+    desk = StubDesk(error=BudgetExhausted("consult budget of 3 is spent"))
+    client = ScriptedClient(
+        [
+            [call("ask_expert", question="again?")],
+            [call("write_file", path="a.py", content="x")],
+            [AssistantText(text="done", model="m")],
+        ]
+    )
+    outcome = build(client, store, ws, expert=desk).run(milestone=1)
+
+    assert outcome.status == "completed"  # it got on with the work
+    results = [
+        e.event for e in store.read_events() if type(e.event).__name__ == "ToolResult"
+    ]
+    assert not results[0].success
+    assert "budget" in results[0].result