Merge pull request 'Threads, commands and labels on the new loop (M2.3)' (#102) from loop-labels into main

Reviewed-on: dongar/nisse#102
Reviewed-by: Don Garrett <dgarrett@acm.org>

docs/Development-Plan.md

@@ -27,7 +27,7 @@ Only `nisse-dev` has real code; `nisse-core` and `nisse-ops` are placeholders.
 | **L2** Agent loop | partial | `control.py`, a fixed state machine with no-stranding, `stop`, `abandon`. Replaced by the new loop (Phase 2) |
 | **L3** Authority | partial | Revision binding and arrival stamping (`gates.py`); scope is still plan approval, not the issue and its conversation |
 | **L4** Classifier | none | |
-| **L5** Conversation | partial | Matrix transport, mentions, lenient commands; no threads, no question handling, little explanation |
+| **L5** Conversation | partial | Matrix transport, mentions, lenient commands; on the new loop, a thread per item, named and ambiguous commands, plain-language go-aheads (M2.3); little explanation |
 | **L6** Context | partial | Context fitting, per-role views and orchestrator facts (M1.2); no scope or agent memory |
 | **L7** Canonical log | partial | Work item and actor on every event, `fact` (M1.1, M1.2); no `classifier_verdict`, `knowledge_retrieved` |
 | **L8** Durability | mostly done | WAL store, intents, boot reconciliation, PR adoption, comment redelivery; to move onto the new loop |
@@ -36,7 +36,7 @@ Only `nisse-dev` has real code; `nisse-core` and `nisse-ops` are placeholders.
 | **L11** Injection | partial | Operator allowlists on every channel; no classifier |
 | **L12** Secrets, logging | partial | Structured state-change log; secrets in `.env`; no per-operation handout or redaction |
 | **L13** Execution | none | Tools run locally in the agent's process. The token leaks into the clone (#89) |
-| **D1–D2** Workflow | none | Plans posted to a Gitea issue and gated; no issue intake, labels, or documents in the repo |
+| **D1–D2** Workflow | partial | On the new loop: issue intake, restatement and go-ahead (M2.1), labels (M2.3); no documents in the repo |
 | **D3–D4** Workspace, git | partial | Idempotent publishing; no fresh workspace (#62), `git add -A` (#80) |
 | **D5** Review | partial | PR conversation comments only (#69), watch stall (#70) |
 | **D7–D8** Deployment, projects | none | Runs from a venv as a systemd unit on this box; one project in config |
@@ -154,9 +154,18 @@ dependencies, a closed PR handled (L1, D4). End-to-end: two issues and a depende
   and checks that parked work is never lost (every file an open item wrote is on
   the workspace or in the forge).
 
-**M2.3 Threads and labels.** *Pre-approved.* Routing by thread, ambiguous commands
-asked about (L5); `bug` grace periods, `discuss`, `design`; questions and
-plain-language go-aheads (D1). End-to-end: a bug and a feature, labelled and not.
+**M2.3 Threads and labels.** *Pre-approved.* **Done** (this PR). Routing by thread,
+ambiguous commands asked about (L5); `bug` grace periods, `discuss`, `design`;
+questions and plain-language go-aheads (D1). End-to-end: a bug and a feature,
+labelled and not.
+- Commands naming their item ("abandon #14") apply to it anywhere; `go`, `approve`
+  and `abandon` alone at top level are asked about. `stop` is never an answer.
+- A plain-language go-ahead is read by one model call (`GoAheadReader`, on the
+  driver's model until the classifier role, M2.6); questions never reach it.
+- `bug` waits `tuning.loop.bug_grace_s` (default 10 minutes) after its restatement;
+  any reply stops the countdown. `discuss` isn't offered `restate` until the
+  operator has spoken; `design` is steered by the driver's instructions.
+- Labels are read at intake; a label changed later is not picked up yet.
 
 **M2.4 Multi-milestone work and documents.** *Pre-approved.* Build plans in the
 repository, `ARCHITECTURE.md` kept current, waiting for merges between milestones

docs/Tuning-Guide.md

@@ -513,3 +513,16 @@ Later work that will add knobs, each with its own section here:
   model and a policy for what is safe to summarize.
 - **Benchmarking the other roles:** planner and delegate, with delegation
   turned back on for comparisons across frontier tiers.
+
+### `tuning.loop.bug_grace_s` — default 600
+
+How long the restatement of an issue labelled `bug` waits for a reply before
+the agent goes ahead on its own (architecture D1). Any reply in the item's
+thread stops the countdown, and `go` starts at once.
+
+- **Too low:** you can't read the restatement before the fix starts, so a
+  misread report costs a wasted milestone.
+- **Too high:** bugs wait as long as features, and the label buys nothing.
+- **Deciding:** about the time it takes you to notice a message and read it
+  when you're around. A wrong start costs only time, since the fix still comes
+  as a PR you review.

nisse-dev.example.yaml

@@ -69,3 +69,5 @@ tuning:
   context:                   # fitting each request into the model's window
     keep_recent_turns: 6     # turns never cut or dropped
     safety_margin: 0.1       # fraction of the window held back
+  loop:                      # the new loop (loop: new)
+    bug_grace_s: 600         # a `bug` restatement proceeds after this long unanswered

nisse-dev/src/nisse_dev/config.py

@@ -104,6 +104,13 @@ class ContextTuning:
     safety_margin: float = 0.1
 
 
+@dataclass(frozen=True)
+class LoopTuning:
+    # A restatement on a `bug` issue proceeds after this long without a
+    # reply (D1).
+    bug_grace_s: float = 600.0
+
+
 @dataclass(frozen=True)
 class Tuning:
     executor: ExecutorTuning = field(default_factory=ExecutorTuning)
@@ -111,6 +118,7 @@ class Tuning:
     delegation: DelegationTuning = field(default_factory=DelegationTuning)
     tools: ToolsTuning = field(default_factory=ToolsTuning)
     context: ContextTuning = field(default_factory=ContextTuning)
+    loop: LoopTuning = field(default_factory=LoopTuning)
 
 
 _TUNING_SECTIONS: dict[str, type] = {

nisse-dev/src/nisse_dev/loop/agent.py

@@ -17,6 +17,7 @@ run (L1).
 
 import re
 import threading
+import time
 import uuid
 from collections.abc import Callable, Iterable
 from dataclasses import replace
@@ -42,6 +43,7 @@ from nisse_dev.loop.items import (
     save,
 )
 from nisse_dev.loop.milestones import OPEN_PR, Milestones
+from nisse_dev.loop.reader import GoAheadReader, is_question
 from nisse_dev.loop.tools import (
     ASK_HUMAN,
     CLOSE_ITEM,
@@ -59,6 +61,11 @@ from nisse_dev.worker import Worker
 INTAKE_INTERVAL_S = 30.0
 _GO = frozenset({"go", "approve"})
 _NEXT = re.compile(r"(?:do|work on) #?(\d+) (?:next|first|now)")
+# A command that names its item applies to it, wherever it is said (L5).
+_NAMED = re.compile(r"(go|approve|abandon|stop) #(\d+)")
+# At top level these need an item; without one they are ambiguous (L5).
+_NEEDS_ITEM = frozenset({"go", "approve", "abandon"})
+BUG_GRACE_S = 600.0
 
 # Before the go-ahead the driver may look around (read, list, run a command
 # to reproduce a bug) but not change anything (D1); after it, everything.
@@ -71,7 +78,8 @@ The issue and the conversation about it are below.
 Before any work, find out what you need, then use restate to put your
 understanding to the operator: what you take the request to mean, how it fits
 the project, and your first step. Wait for their go-ahead. If the request is
-unclear, ask_human instead of guessing.
+unclear, ask_human instead of guessing: a restatement asks no questions, so
+the operator can agree to it as it stands.
 
 After the go-ahead, work in a milestone: call start_milestone before
 changing anything (never create or switch branches yourself), make the
@@ -84,8 +92,25 @@ the PR is closed without merging, ask the operator what next. When a review
 needs no change from you, or you are only waiting for the operator to merge
 (you can't merge it yourself), call wait_for_review.
 
+{labels}Never label the issue yourself; you may suggest a label to the operator.
+
 {workspace}"""
 
+# What an issue's labels add to the driver's instructions (D1).
+LABELS = {
+    "bug": "This issue is labelled bug. Before restating, reproduce it with "
+    "your tools (run the code or the tests). If you can't reproduce it, "
+    "ask_human instead of restating: never fix a bug you couldn't see. Once "
+    "you restate, the fix goes ahead after a grace period unless the operator "
+    "replies; make it a failing test first, then the fix.",
+    "discuss": "This issue is labelled discuss: talk it through with the "
+    "operator (ask_human) before restating, and restate only once you agree "
+    "on what is wanted.",
+    "design": "This issue is labelled design: agree the design with the "
+    "operator before any code. Your restatement puts the design to them; "
+    "code waits for their go-ahead on it.",
+}
+
 
 class LoopAgent:
     def __init__(
@@ -106,6 +131,9 @@ class LoopAgent:
         oplog: OpLog,
         roles: dict[str, str] | None = None,
         milestones: Milestones | None = None,
+        reader: GoAheadReader | None = None,
+        bug_grace_s: float = BUG_GRACE_S,
+        clock: Callable[[], float] = time.time,
         intake_interval_s: float = INTAKE_INTERVAL_S,
     ) -> None:
         self._store = store
@@ -123,10 +151,16 @@ class LoopAgent:
         self._oplog = oplog
         self._roles = roles or {}
         self._milestones = milestones
+        self._reader = reader
+        self._grace = bug_grace_s
+        self._clock = clock
         self._interval = intake_interval_s
         # thread root -> revision on screen there, read on the transport's
-        # thread to stamp replies on arrival (L3).
+        # thread to stamp replies on arrival (L3); issue -> thread root, for
+        # a command that names its item.
         self._shown: dict[str, int] = {}
+        self._threads: dict[int, str] = {}
+        self._active_thread = ""
         self._running = False
         self._stop = threading.Event()
         self._intake_thread: threading.Thread | None = None
@@ -173,6 +207,7 @@ class LoopAgent:
         self._reconcile_publishing()
         for item in self._store.work_items("open"):
             loop = loop_of(item)
+            self._threads[loop.issue] = loop.thread
             if item.position == RUNNING:
                 note = "I restarted while working on this. Reply to continue."
                 self._wait(item, replace(loop, wait=ANSWER, note=note), note)
@@ -249,6 +284,7 @@ class LoopAgent:
             self._open(issue, tuple(blocked_by))
         self.watch()
         self.check_blocked()
+        self.check_grace()
         self._schedule()
 
     # -- PR review and merge (D4, D5) -------------------------------------------
@@ -354,6 +390,24 @@ class LoopAgent:
                 cause=said,
             )
 
+    def check_grace(self) -> None:
+        """D1: a `bug` restatement left unanswered for the grace period is
+        taken as agreed, bound to the revision that was up all along."""
+        now = self._clock()
+        for item in self._store.work_items("open"):
+            loop = loop_of(item)
+            if (
+                item.position == WAITING
+                and loop.wait == GO_AHEAD
+                and loop.grace_until
+                and now >= loop.grace_until
+            ):
+                self._transport.post(
+                    f"No reply on rev-{loop.rev}, so I'm going ahead.",
+                    thread=loop.thread,
+                )
+                self._go(item, loop, "the bug grace period")
+
     def _from_an_operator(self, issue: Issue) -> bool:
         """L11: only an operator's issue, or one an operator assigned, is
         an instruction."""
@@ -375,7 +429,13 @@ class LoopAgent:
         root = self._transport.post(
             f"Queued {self._repo}#{issue.number}: {issue.title}{where}"
         )
-        loop = Loop(thread=root, issue=issue.number, blocked_by=blocked_by)
+        loop = Loop(
+            thread=root,
+            issue=issue.number,
+            blocked_by=blocked_by,
+            labels=tuple(issue.labels),
+        )
+        self._threads[issue.number] = root
         self._store.open_work_item(
             item_id,
             title=issue.title,
@@ -408,10 +468,21 @@ class LoopAgent:
         """On the transport's thread: stamp, then hand to the worker."""
         if msg.sender not in self._operators:
             return
-        if command_form(msg.text) == "stop" and self._running:
+        command = command_form(msg.text)
+        thread = msg.thread
+        named = _NAMED.fullmatch(command)
+        if named is not None:
+            command = named.group(1)
+            thread = self._threads.get(int(named.group(2)), thread)
+        if (
+            command == "stop"
+            and self._running
+            # Unnamed at top level, it stops the active item (L5).
+            and (not thread or thread == self._active_thread)
+        ):
             # Straight to the flag: the run is what it has to reach (L2).
             self._cancel.set()
-        about = self._shown.get(msg.thread) if msg.thread else None
+        about = self._shown.get(thread) if thread else None
         self._worker.submit(
             self._settled("your message", lambda: self.handle(msg, about))
         )
@@ -430,13 +501,38 @@ class LoopAgent:
             )
             self._schedule()
             return
+        named = _NAMED.fullmatch(command)
+        if named is not None:
+            # L5: a command that names its item applies to it.
+            command, issue = named.group(1), int(named.group(2))
+            item = next(
+                (
+                    i
+                    for i in self._store.work_items("open")
+                    if loop_of(i).issue == issue
+                ),
+                None,
+            )
+            if item is None:
+                self._store.transcript(None, "operator").append(
+                    UserMessage(text=text, source=f"transport:{msg.sender}")
+                )
+                self._transport.post(
+                    f"#{issue} isn't one of my open work items.", thread=msg.thread
+                )
+                return
         if item is None:
             self._store.transcript(None, "operator").append(
                 UserMessage(text=text, source=f"transport:{msg.sender}")
             )
             if command == "status":
                 self._transport.post(self.status(), thread=msg.thread)
-            elif command != "stop":
+            elif command in _NEEDS_ITEM:
+                self._transport.post(self._which(command), thread=msg.thread)
+            elif command == "stop":
+                if not self._store.active_item():
+                    self._transport.post("Nothing is running.", thread=msg.thread)
+            else:
                 self._transport.post(
                     f"I take work from issues assigned to me in {self._repo}: "
                     "assign one and I'll pick it up. Reply in its thread to "
@@ -447,12 +543,27 @@ class LoopAgent:
         self._store.transcript(item.id, "operator").append(
             UserMessage(text=text, source=f"transport:{msg.sender}")
         )
-        loop = replace(loop_of(item), asked_on_pr=0, asked_where="", asked_text="")
+        # Any reply stops a grace period's countdown: the operator is here.
+        loop = replace(
+            loop_of(item),
+            asked_on_pr=0,
+            asked_where="",
+            asked_text="",
+            grace_until=0.0,
+            talked=True,
+        )
         if loop != loop_of(item):
             save(self._store, item.id, item.position, loop, cause="operator spoke")
             item = self._store.work_item(item.id) or item
         if command == "abandon":
             self._close(item, loop, "abandoned", f"abandoned by {msg.sender}")
+        elif command == "stop":
+            # Stopping a run took effect on arrival; a stop is never an answer.
+            if not loop.note.startswith("stopped ("):
+                self._transport.post(
+                    f"#{loop.issue} isn't running; nothing to stop.",
+                    thread=loop.thread,
+                )
         elif item.position == BLOCKED:
             self._transport.post(
                 f"#{loop.issue} is blocked by {', '.join(loop.blocked_by)}; "
@@ -468,14 +579,18 @@ class LoopAgent:
                     thread=loop.thread,
                 )
                 return
-            self._fact(item, f"go-ahead on rev-{loop.rev} from {msg.sender}")
-            save(
-                self._store,
-                item.id,
-                READY,
-                replace(loop, started=True, wait=""),
-                cause=f"go-ahead on rev-{loop.rev}",
+            self._go(item, loop, msg.sender)
+        elif (
+            item.position == WAITING
+            and loop.wait == GO_AHEAD
+            and about == loop.rev
+            and self._plain_go(item, text)
+        ):
+            # L5, D1: a plain-language yes, said so before acting.
+            self._transport.post(
+                f"Taking that as a go-ahead on rev-{loop.rev}.", thread=loop.thread
             )
+            self._go(item, loop, f"{msg.sender} (in plain language)")
         elif item.position == WAITING:
             # An answer, or changes to a restatement: the driver reads it.
             save(
@@ -487,6 +602,39 @@ class LoopAgent:
             )
         self._schedule()
 
+    def _go(self, item: WorkItem, loop: Loop, by: str) -> None:
+        self._fact(item, f"go-ahead on rev-{loop.rev} from {by}")
+        save(
+            self._store,
+            item.id,
+            READY,
+            replace(loop, started=True, wait="", grace_until=0.0),
+            cause=f"go-ahead on rev-{loop.rev}",
+        )
+
+    def _plain_go(self, item: WorkItem, text: str) -> bool:
+        """Whether a reply that isn't a command agrees to the restatement up.
+        Questions never do (L5)."""
+        if self._reader is None or is_question(text):
+            return False
+        proposals = [
+            e.event.plan
+            for e in self._store.transcript(item.id, "driver").read()
+            if isinstance(e.event, PlanSubmitted)
+        ]
+        return bool(proposals) and self._reader.agrees(proposals[-1], text)
+
+    def _which(self, command: str) -> str:
+        """L5: an unnamed command at top level is ambiguous: ask."""
+        items = self._store.work_items("open")
+        if not items:
+            return f"{command} what? I have no open work items."
+        listed = ", ".join(f"#{loop_of(i).issue} ({i.title})" for i in items)
+        return (
+            f"Which item? Say it in the item's thread, or name it "
+            f'("{command} #{loop_of(items[0]).issue}"). Open: {listed}.'
+        )
+
     def _put_first(self, issue: int) -> str:
         """The operator's "do #N next" (L1): it runs before every other
         ready item, the latest such order first. It doesn't unblock."""
@@ -525,6 +673,9 @@ class LoopAgent:
             what = item.position
             if item.position == WAITING:
                 what = f"waiting for {loop.wait} ({loop.note or f'rev-{loop.rev}'})"
+                if loop.grace_until:
+                    left = max(0, int(loop.grace_until - self._clock()) // 60)
+                    what += f", going ahead in {left} min unless you reply"
             elif item.position == BLOCKED:
                 what = f"blocked by {', '.join(loop.blocked_by)}"
             elif item.position == READY:
@@ -574,8 +725,10 @@ class LoopAgent:
                 self._transport.post(said, thread=loop.thread)
             loop = replace(loop, prepared=True)
         save(self._store, item.id, RUNNING, loop, cause="driver run")
+        # D1: a `discuss` item talks before it restates.
+        hold = "discuss" in loop.labels and not loop.talked and not loop.started
         tools = [
-            RESTATE,
+            *([] if hold else [RESTATE]),
             ASK_HUMAN,
             *(
                 _LOOK
@@ -593,8 +746,10 @@ class LoopAgent:
             repo=self._repo,
             issue=loop.issue,
             title=item.title,
+            labels="".join(f"{LABELS[x]}\n\n" for x in loop.labels if x in LABELS),
             workspace=self._workspace.describe() if self._workspace else "",
         )
+        self._active_thread = loop.thread
         self._running = True
         try:
             outcome = self._driver.run(
@@ -622,11 +777,19 @@ class LoopAgent:
             )
             self._fact(item, f"restatement rev-{rev} put up for a go-ahead")
             self._shown[loop.thread] = rev
-            self._transport.post(
-                f"rev-{rev}: {text}\n\nSay go to start, or reply with changes.",
-                thread=loop.thread,
+            grace = self._clock() + self._grace if "bug" in loop.labels else 0.0
+            after = "Say go to start, or reply with changes."
+            if grace:
+                after = (
+                    f"It's a bug, so I'll go ahead in {round(self._grace / 60)} min "
+                    "unless you reply; say go to start now, or reply with changes."
+                )
+            self._transport.post(f"rev-{rev}: {text}\n\n{after}", thread=loop.thread)
+            self._wait(
+                item,
+                replace(loop, rev=rev, wait=GO_AHEAD, note="", grace_until=grace),
+                None,
             )
-            self._wait(item, replace(loop, rev=rev, wait=GO_AHEAD, note=""), None)
         elif outcome.kind == "wait" and outcome.tool == "ask_human":
             question = str(args.get("question", "")).strip()
             self._answer_where_asked(loop, question)

nisse-dev/src/nisse_dev/loop/build.py

@@ -19,6 +19,7 @@ from nisse_dev.gitea import GiteaClient
 from nisse_dev.loop.agent import LoopAgent
 from nisse_dev.loop.driver import Driver
 from nisse_dev.loop.milestones import Milestones
+from nisse_dev.loop.reader import GoAheadReader
 from nisse_dev.matrix import MatrixTransport
 from nisse_dev.oplog import OpLog
 from nisse_dev.state import StateStore
@@ -109,6 +110,10 @@ def build_loop_agent(config: AgentConfig) -> RunningLoop:
         cancel=cancel,
         oplog=oplog,
         roles=config.roles.pinned(),
+        # The driver's model reads plain-language go-aheads until the
+        # classifier role arrives (M2.6).
+        reader=GoAheadReader(normalized, config.roles.resolve("planner")),
+        bug_grace_s=config.tuning.loop.bug_grace_s,
         milestones=Milestones(
             store,
             git,

nisse-dev/src/nisse_dev/loop/items.py

@@ -40,6 +40,9 @@ class Loop:
     fixes: int = 0  # review fixes pushed, for follow-up branch names
     blocked_by: tuple[str, ...] = ()  # open dependencies, "owner/repo#N" (L1)
     first: int = 0  # the operator's "do #N next": higher runs sooner
+    labels: tuple[str, ...] = ()  # the issue's labels at intake: they steer (D1)
+    grace_until: float = 0.0  # a `bug` restatement proceeds then, unanswered
+    talked: bool = False  # the operator has spoken in the item's thread
     # Where the operator last spoke (D5: answer where asked): "" for the
     # Matrix thread, or the PR it came from, with the inline location and a
     # little of what they said, for quoting.
@@ -55,6 +58,7 @@ def loop_of(item: WorkItem) -> Loop:
     data = dict(item.data.get("loop", {}))
     data["notes"] = tuple(data.get("notes", ()))
     data["blocked_by"] = tuple(data.get("blocked_by", ()))
+    data["labels"] = tuple(data.get("labels", ()))
     return Loop(**data)
 
 

nisse-dev/src/nisse_dev/loop/reader.py

@@ -0,0 +1,81 @@
+"""Reading a plain-language reply as a go-ahead (L5, D1).
+
+"Yep, that's it" in reply to a restatement may be taken as `go`, because a
+wrong reading costs only time. One short model call decides, and it decides
+conservatively: only a clear yes counts. A question is never a go-ahead and
+never reaches the model (L5: questions never change state).
+
+The driver's model reads for now; the classifier role takes over in M2.6.
+"""
+
+from typing import Any
+
+from nisse_dev.canonical import StoredEvent, ToolCall, UserMessage
+from nisse_dev.llm import LLMClient, LLMError
+
+VERDICT = {
+    "type": "function",
+    "function": {
+        "name": "verdict",
+        "description": "Your reading of the operator's reply.",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "answer": {
+                    "type": "string",
+                    "enum": ["yes", "no", "unclear"],
+                    "description": "yes: a clear go-ahead on the proposal as it "
+                    "stands. no: it asks for changes, disagrees or says to wait. "
+                    "unclear: anything else.",
+                }
+            },
+            "required": ["answer"],
+        },
+    },
+}
+
+SYSTEM = """You read one reply from an operator to a software agent. The agent
+put a proposal up for agreement; decide whether the reply clearly agrees to the
+proposal as it stands, so the agent may start. A reply that asks for any
+change, asks a question, adds a condition, or is about something else is not a
+go-ahead. When in doubt, it is not. Answer with the verdict tool."""
+
+
+def is_question(text: str) -> bool:
+    return text.rstrip().endswith("?")
+
+
+class GoAheadReader:
+    def __init__(self, client: LLMClient, model: str) -> None:
+        self._client = client
+        self._model = model
+
+    def agrees(self, proposal: str, reply: str) -> bool:
+        if is_question(reply):
+            return False
+        prompt = f"The proposal:\n\n{proposal}\n\nThe operator's reply:\n\n{reply}"
+        try:
+            produced = self._client.complete(
+                [
+                    StoredEvent(
+                        id=0,
+                        ts="",
+                        event=UserMessage(text=prompt, source="go-ahead reader"),
+                        actor="orchestrator",
+                    )
+                ],
+                tools=[VERDICT],
+                model=self._model,
+                system=SYSTEM,
+            )
+        except LLMError:
+            return False  # unread: the driver reads it as an answer instead
+        return any(_yes(event) for event in produced)
+
+
+def _yes(event: Any) -> bool:
+    return (
+        isinstance(event, ToolCall)
+        and event.tool == "verdict"
+        and event.arguments.get("answer") == "yes"
+    )

nisse-dev/tests/test_config.py

@@ -158,6 +158,7 @@ def test_tuning_is_read() -> None:
             "expert": {"consult_budget": 5},
             "delegation": {"max_turns": 20, "max_tokens": 100000, "wall_clock_s": 600},
             "tools": {"command_timeout_s": 300, "max_output_chars": 2000},
+            "loop": {"bug_grace_s": 120},
         }
     )
     tuning = config.tuning
@@ -166,6 +167,7 @@ def test_tuning_is_read() -> None:
     assert tuning.delegation.max_tokens == 100_000
     assert tuning.delegation.wall_clock_s == 600.0
     assert tuning.tools.max_output_chars == 2000
+    assert tuning.loop.bug_grace_s == 120.0
 
 
 def test_a_partly_given_tuning_section_keeps_the_other_defaults() -> None:

nisse-dev/tests/test_loop.py

@@ -72,12 +72,18 @@ def issue(number: int = 14, author: str = "don", **kw: Any) -> Issue:
         title=kw.get("title", "Add count_words"),
         body=kw.get("body", "count words in a string"),
         author=author,
-        labels=(),
+        labels=kw.get("labels", ()),
     )
 
 
 class Harness:
-    def __init__(self, tmp_path: Path, model: ScriptedModel, gitea: FakeGitea) -> None:
+    def __init__(
+        self,
+        tmp_path: Path,
+        model: ScriptedModel,
+        gitea: FakeGitea,
+        **agent: Any,  # more LoopAgent arguments: reader, clock, ...
+    ) -> None:
         tmp_path.mkdir(parents=True, exist_ok=True)
         self.store = StateStore(
             tmp_path / "state.db", oplog=OpLog("nisse-dev", io.StringIO())
@@ -100,6 +106,7 @@ class Harness:
             worker=InlineWorker(),
             cancel=self.cancel,
             oplog=OpLog("nisse-dev", io.StringIO()),
+            **agent,
         )
         self.transport.on_message(self.agent.on_message)
 
@@ -215,7 +222,8 @@ def test_a_go_at_top_level_is_not_a_go_ahead(h: Harness) -> None:
     h.agent.intake()
     h.say("go", thread=None)
     assert not loop_of(h.item()).started
-    assert "assign one and I'll pick it up" in h.posted()[-1]
+    # L5: an unnamed command at top level is ambiguous, so it asks.
+    assert "Which item?" in h.posted()[-1] and "#14" in h.posted()[-1]
 
 
 # --- waiting points (L2) -------------------------------------------------------------

nisse-dev/tests/test_loop_labels.py

@@ -0,0 +1,282 @@
+"""Threads, commands and labels in the new loop (M2.3): commands that name
+their item, ambiguous ones asked about (L5); plain-language go-aheads; `bug`
+grace periods, `discuss` and `design` (D1)."""
+
+from pathlib import Path
+from typing import Any
+
+import pytest
+from test_loop import DON, RESTATED, FakeGitea, Harness, ScriptedModel, call, issue
+from test_loop import h as h  # noqa: PLC0414 -- the fixture
+
+from nisse_dev.canonical import Fact, ToolCall
+from nisse_dev.llm import LLMError
+from nisse_dev.loop.items import ANSWER, GO_AHEAD, READY, WAITING, loop_of
+from nisse_dev.loop.reader import GoAheadReader
+
+STARTED = [call("ask_human", question="started; anything else?")]
+
+
+class Clock:
+    def __init__(self) -> None:
+        self.now = 1000.0
+
+    def __call__(self) -> float:
+        return self.now
+
+
+class Reader:
+    """A scripted go-ahead reader: agrees with what it's told to."""
+
+    def __init__(self, *yes: str) -> None:
+        self.yes = set(yes)
+        self.asked: list[tuple[str, str]] = []
+
+    def agrees(self, proposal: str, reply: str) -> bool:
+        self.asked.append((proposal, reply))
+        return reply in self.yes
+
+
+def facts(h: Harness) -> list[str]:
+    return [
+        e.event.text
+        for e in h.store.transcript(h.item().id, "orchestrator").read()
+        if isinstance(e.event, Fact)
+    ]
+
+
+def two(tmp_path: Path, *turns: list[Any]) -> Harness:
+    return Harness(
+        tmp_path,
+        ScriptedModel(*turns),
+        FakeGitea(issue(14), issue(15, title="Add count_lines")),
+    )
+
+
+# --- commands that name their item, and ambiguous ones (L5) -------------------
+
+
+def test_a_named_go_at_top_level_applies_to_its_item(tmp_path: Path) -> None:
+    h = two(tmp_path, RESTATED, RESTATED, STARTED)
+    h.agent.intake()
+    h.say("go #15", thread=None)
+    items = {loop_of(i).issue: i for i in h.store.work_items()}
+    assert loop_of(items[15]).started and not loop_of(items[14]).started
+
+
+def test_a_named_abandon_in_another_thread_still_names_its_item(
+    tmp_path: Path,
+) -> None:
+    h = two(tmp_path, RESTATED, RESTATED)
+    h.agent.intake()
+    items = {loop_of(i).issue: i for i in h.store.work_items()}
+    h.transport.receive(DON, "Abandon #15.", thread=loop_of(items[14]).thread)
+    items = {loop_of(i).issue: i for i in h.store.work_items()}
+    assert items[15].status == "closed" and items[14].status == "open"
+
+
+def test_an_unnamed_command_at_top_level_is_asked_about(tmp_path: Path) -> None:
+    h = two(tmp_path, RESTATED, RESTATED)
+    h.agent.intake()
+    h.say("abandon", thread=None)
+    assert all(i.status == "open" for i in h.store.work_items())
+    asked = h.posted()[-1]
+    assert "Which item?" in asked and "#14" in asked and "#15" in asked
+
+
+def test_naming_an_item_that_isnt_open_says_so(h: Harness) -> None:
+    h.agent.intake()
+    h.say("go #99", thread=None)
+    assert "#99 isn't one of my open work items" in h.posted()[-1]
+
+
+def test_stop_in_a_waiting_items_thread_is_not_an_answer(h: Harness) -> None:
+    h.agent.intake()
+    h.say("stop")
+    assert (h.item().position, loop_of(h.item()).wait) == (WAITING, GO_AHEAD)
+    assert "nothing to stop" in h.posted()[-1]
+
+
+# --- plain-language go-aheads (L5, D1) ------------------------------------------
+
+
+def test_a_plain_yes_is_a_go_ahead_said_so_first(tmp_path: Path) -> None:
+    reader = Reader("Yep, that's it")
+    h = Harness(
+        tmp_path, ScriptedModel(RESTATED, STARTED), FakeGitea(issue()), reader=reader
+    )
+    h.agent.intake()
+    h.say("Yep, that's it")
+    assert loop_of(h.item()).started
+    assert "Taking that as a go-ahead on rev-1" in " ".join(h.posted())
+    assert reader.asked == [
+        ("I'll add count_words to text.py, with tests.", "Yep, that's it")
+    ]
+    assert any(f.startswith("go-ahead on rev-1 from") for f in facts(h))
+
+
+def test_anything_else_goes_to_the_driver(tmp_path: Path) -> None:
+    h = Harness(
+        tmp_path,
+        ScriptedModel(RESTATED, [call("restate", text="v2")]),
+        FakeGitea(issue()),
+        reader=Reader(),
+    )
+    h.agent.intake()
+    h.say("Mostly, but call it word_count")
+    loop = loop_of(h.item())
+    assert not loop.started and loop.rev == 2
+
+
+def test_a_question_never_reaches_the_reader(tmp_path: Path) -> None:
+    reader = Reader("sounds good?")
+    h = Harness(
+        tmp_path,
+        ScriptedModel(RESTATED, [call("restate", text="v2")]),
+        FakeGitea(issue()),
+        reader=reader,
+    )
+    h.agent.intake()
+    h.say("sounds good?")
+    assert not loop_of(h.item()).started and reader.asked == []
+
+
+def test_a_plain_yes_to_an_older_revision_is_not_a_go(tmp_path: Path) -> None:
+    reader = Reader("yes")
+    h = Harness(
+        tmp_path,
+        ScriptedModel(RESTATED, [call("restate", text="v2")], STARTED),
+        FakeGitea(issue()),
+        reader=reader,
+    )
+    h.agent.intake()
+    h.say("change it a bit")  # rev-2 goes up
+    h.agent.handle(_msg("yes", h.thread()), about=1)
+    assert not loop_of(h.item()).started
+    assert [reply for _, reply in reader.asked] == ["change it a bit"]
+
+
+class _Model:
+    def __init__(self, result: Any) -> None:
+        self.result = result
+
+    def complete(self, events, tools, model, system=None) -> list[Any]:
+        if isinstance(self.result, Exception):
+            raise self.result
+        return self.result
+
+
+@pytest.mark.parametrize(
+    ("result", "agrees"),
+    [
+        ([ToolCall(call_id="v", tool="verdict", arguments={"answer": "yes"})], True),
+        (
+            [ToolCall(call_id="v", tool="verdict", arguments={"answer": "unclear"})],
+            False,
+        ),
+        ([], False),
+        (LLMError("down"), False),
+    ],
+)
+def test_the_reader_agrees_only_on_a_clear_yes(result: Any, agrees: bool) -> None:
+    reader = GoAheadReader(_Model(result), "m")  # type: ignore[arg-type]
+    assert reader.agrees("add count_words", "ok") is agrees
+
+
+# --- labels (D1) ----------------------------------------------------------------
+
+
+def bug(tmp_path: Path, *turns: list[Any], **kw: Any) -> tuple[Harness, Clock]:
+    clock = Clock()
+    h = Harness(
+        tmp_path,
+        ScriptedModel(*turns),
+        FakeGitea(issue(labels=("bug",))),
+        clock=clock,
+        bug_grace_s=600,
+        **kw,
+    )
+    return h, clock
+
+
+def test_a_bug_goes_ahead_after_the_grace_period(tmp_path: Path) -> None:
+    h, clock = bug(tmp_path, RESTATED, STARTED)
+    h.agent.intake()
+    assert "go ahead in 10 min unless you reply" in h.posted()[-1]
+    assert "going ahead in 10 min" in h.agent.status()
+    clock.now += 599
+    h.agent.intake()
+    assert not loop_of(h.item()).started
+    clock.now += 1
+    h.agent.intake()
+    assert loop_of(h.item()).started
+    assert "go-ahead on rev-1 from the bug grace period" in facts(h)
+
+
+def test_a_reply_stops_the_bugs_countdown(tmp_path: Path) -> None:
+    h, clock = bug(tmp_path, RESTATED, [call("ask_human", question="which file?")])
+    h.agent.intake()
+    h.say("hmm, what about unicode")
+    clock.now += 3600
+    h.agent.intake()
+    loop = loop_of(h.item())
+    assert not loop.started and loop.grace_until == 0 and loop.wait == ANSWER
+
+
+def test_a_new_revision_of_a_bug_gets_its_own_countdown(tmp_path: Path) -> None:
+    h, clock = bug(tmp_path, RESTATED, [call("restate", text="v2")], STARTED)
+    h.agent.intake()
+    clock.now += 300
+    h.say("use the other file")
+    assert loop_of(h.item()).grace_until == clock.now + 600
+    clock.now += 600
+    h.agent.intake()
+    assert "go-ahead on rev-2 from the bug grace period" in facts(h)
+
+
+def test_the_bug_label_reaches_the_driver(tmp_path: Path) -> None:
+    seen: list[str] = []
+
+    class Spy(ScriptedModel):
+        def complete(self, events, tools, model, system=None):  # type: ignore[no-untyped-def]
+            seen.append(system or "")
+            return super().complete(events, tools, model, system)
+
+    h = Harness(tmp_path, Spy(RESTATED), FakeGitea(issue(labels=("bug", "ui"))))
+    h.agent.intake()
+    assert "never fix a bug you couldn't see" in seen[0]
+    assert "labelled discuss" not in seen[0]
+
+
+def test_discuss_talks_before_it_restates(tmp_path: Path) -> None:
+    offered: list[set[str]] = []
+
+    class Spy(ScriptedModel):
+        def complete(self, events, tools, model, system=None):  # type: ignore[no-untyped-def]
+            offered.append({t["function"]["name"] for t in tools})
+            return super().complete(events, tools, model, system)
+
+    h = Harness(
+        tmp_path,
+        Spy([call("ask_human", question="what's it for?")], RESTATED),
+        FakeGitea(issue(labels=("discuss",))),
+    )
+    h.agent.intake()
+    assert "restate" not in offered[0]
+    assert loop_of(h.item()).wait == ANSWER
+    h.say("for the stats page")
+    assert "restate" in offered[1]
+    assert loop_of(h.item()).wait == GO_AHEAD
+
+
+def test_an_unlabelled_issue_waits_for_go_with_no_countdown(h: Harness) -> None:
+    h.agent.intake()
+    loop = loop_of(h.item())
+    assert loop.wait == GO_AHEAD and loop.grace_until == 0
+    assert h.item().position != READY
+
+
+def _msg(text: str, thread: str) -> Any:
+    from nisse_dev.transport import InboundMessage
+
+    return InboundMessage(DON, text, thread=thread)

nisse-dev/tests/test_loop_properties.py

@@ -19,7 +19,7 @@ from test_loop_milestones import DON, Forge, call
 from test_loop_milestones import world as world  # noqa: PLC0414 -- the fixture
 
 from nisse_dev.cancel import Cancel
-from nisse_dev.canonical import Fact, ToolCall
+from nisse_dev.canonical import Fact, PlanSubmitted, ToolCall, UserMessage
 from nisse_dev.gitea import Comment, Issue
 from nisse_dev.loop.agent import LoopAgent
 from nisse_dev.loop.driver import Driver
@@ -60,6 +60,8 @@ class PolicyModel:
             return [call("ask_human", question=f"question {self.n}?")]
         if roll < 0.12:
             return [ToolCall(call_id=f"p{self.n}", tool="nonsense", arguments={})]
+        if "restate" not in offered and "finish_milestone" not in offered:
+            return [call("ask_human", question=f"talk first {self.n}?")]  # discuss
         if "finish_milestone" not in offered:
             return [call("restate", text=f"restatement {self.n}")]
         if "wait_for_review" in offered and self.rng.random() < 0.2:
@@ -89,8 +91,14 @@ class Walk:
         self.steps: list[str] = []
         self.shown_at_go: list[str] = []  # item:revision on screen at each go
         self.next_issue = 15
+        self.now = 1000.0
         self.agent, self.transport = self._agent()
 
+    def reads_yes(self, proposal: str, reply: str) -> bool:
+        """The scripted go-ahead reader: a plain yes usually reads as one,
+        and nothing else ever does."""
+        return reply == "yep, that's right" and self.rng.random() < 0.7
+
     def _agent(self) -> tuple[LoopAgent, InMemoryTransport]:
         transport = InMemoryTransport()
         cancel = Cancel()
@@ -110,10 +118,16 @@ class Walk:
             cancel=cancel,
             oplog=OpLog("nisse-dev", _Null()),
             milestones=w["milestones"],
+            reader=self,  # type: ignore[arg-type]
+            bug_grace_s=600,
+            clock=lambda: self.now,
         )
         transport.on_message(agent.on_message)
         return agent, transport
 
+    def agrees(self, proposal: str, reply: str) -> bool:
+        return self.reads_yes(proposal, reply)
+
     def items(self) -> list[Any]:
         return self.world["store"].work_items()
 
@@ -127,7 +141,7 @@ class Walk:
         loop = loop_of(item) if item else None
         if item and loop:
             weights.update({"go": 3, "answer": 2, "abandon": 1, "go-stale": 1})
-            weights.update({"first": 1})
+            weights.update({"first": 1, "plain-yes": 2, "named-go": 1, "wait": 2})
             if loop.pr:
                 weights.update(
                     {"comment": 4, "merge": 2, "close-pr": 1, "comment-merge": 2}
@@ -139,11 +153,33 @@ class Walk:
         elif action == "issue":
             # A new issue, sometimes depending on one that is still open.
             n, self.next_issue = self.next_issue, self.next_issue + 1
-            forge.issues.append(Issue(n, f"task {n}", f"do task {n}", "don", ()))
+            labels = self.rng.choice([(), (), ("bug",), ("discuss",), ("design",)])
+            forge.issues.append(Issue(n, f"task {n}", f"do task {n}", "don", labels))
             live = [i.number for i in forge.assigned_issues("", "") if i.number != n]
             if live and self.rng.random() < 0.5:
                 forge.deps[n] = [self.rng.choice(live)]
             self.agent.intake()
+        elif action == "plain-yes" and loop:
+            # Read as a go-ahead or not; either way bound to what's up now.
+            if item.position == WAITING and loop.wait == GO_AHEAD:
+                self.shown_at_go.append(f"{item.id}:{loop.rev}")
+            self.transport.receive(DON, "yep, that's right", thread=loop.thread)
+        elif action == "named-go" and loop:
+            if item.position == WAITING and loop.wait == GO_AHEAD:
+                self.shown_at_go.append(f"{item.id}:{loop.rev}")
+            self.transport.receive(DON, f"go #{loop.issue}")
+        elif action == "wait":
+            # Time passes: bug countdowns may run out.
+            self.now += self.rng.choice([60, 700])
+            for other in open_items:
+                o = loop_of(other)
+                if (
+                    other.position == WAITING
+                    and o.wait == GO_AHEAD
+                    and 0 < o.grace_until <= self.now
+                ):
+                    self.shown_at_go.append(f"{other.id}:{o.rev}")
+            self.agent.intake()
         elif action == "first" and loop:
             self.agent.handle(InboundMessage(DON, f"do #{loop.issue} next"), None)
         elif action == "crash":
@@ -213,7 +249,7 @@ class Walk:
                 and f.event.text.startswith("go-ahead on rev-")
             ]
             assert all(f"{item.id}:{rev}" in self.shown_at_go for rev in gos), (
-                f"go_binds: {gos}: {where}"
+                f"go_binds: {item.id} {gos} not in {self.shown_at_go}: {where}"
             )
             # Blocked comes from external truth (L1): an item whose issue
             # had an open dependency from the start never ran before it closed.
@@ -238,6 +274,36 @@ class Walk:
                 assert not [
                     i for i in store.pending_intents() if i.work_item == item.id
                 ], f"closed_items_leave_nothing (intent): {where}"
+        for item in items:
+            # D1: a bug's grace period runs out only on an unanswered
+            # restatement: any reply since it went up stops the countdown.
+            replied = False
+            for e in store.transcript(item.id, "orchestrator").read():
+                text = getattr(e.event, "text", "")
+                if isinstance(e.event, Fact) and text.startswith("restatement rev-"):
+                    replied = False
+                elif isinstance(e.event, UserMessage) and e.actor == "operator":
+                    replied = replied or e.event.source.startswith("transport:")
+                elif text.endswith("from the bug grace period"):
+                    assert not replied, (
+                        f"grace_only_unanswered: #{loop_of(item).issue}: {where}"
+                    )
+        for item in items:
+            # D1: a `discuss` item talks before it restates.
+            if "discuss" not in loop_of(item).labels:
+                continue
+            events = store.transcript(item.id, "driver").read()
+            plans = [e.id for e in events if isinstance(e.event, PlanSubmitted)]
+            spoke = [
+                e.id
+                for e in events
+                if e.actor == "operator"
+                and isinstance(e.event, UserMessage)
+                and e.event.source.startswith("transport:")
+            ]
+            assert not plans or (spoke and spoke[0] < plans[0]), (
+                f"discuss_talks_first: #{loop_of(item).issue}: {where}"
+            )
         assert not [i for i in items if i.position == RUNNING], (
             f"one_active (nothing left running): {where}"
         )

tests/integration/test_end_to_end_loop.py

@@ -9,6 +9,7 @@ the item's Matrix thread through the in-memory transport. Run with
 scripts/e2e.sh after each milestone.
 """
 
+import base64
 import contextlib
 import os
 import time
@@ -22,7 +23,7 @@ from typing import Any
 import pytest
 from e2e_support import base_branch, dependent_task, operator_gitea, task
 
-from nisse_dev.config import AgentConfig
+from nisse_dev.config import AgentConfig, LoopTuning
 from nisse_dev.execution import SSHHost
 from nisse_dev.gitea import GiteaClient, GiteaError
 from nisse_dev.loop.build import RunningLoop, build_loop_agent
@@ -42,6 +43,7 @@ pytestmark = [pytest.mark.integration, pytest.mark.e2e]
 OPERATOR = "@dongar:dongar.net"  # an allowlisted Matrix operator in nisse-test.yaml
 DEADLINE_S = 1200
 MAX_ANSWERS = 5  # per wait for something else, before giving up
+BUG_GRACE_S = 20
 
 
 @pytest.fixture
@@ -69,6 +71,8 @@ def running(
         state_db=keep / "state.db",
         workspace=Path(f"/tmp/nisse-e2e-{run}/sandbox"),
         build_host=build_host(ssh_host),
+        # Short, so a bug's grace period runs out within the test.
+        tuning=replace(live_config.tuning, loop=LoopTuning(bug_grace_s=BUG_GRACE_S)),
     )
     transport = InMemoryTransport()
     monkeypatch.setattr("nisse_dev.loop.build.MatrixTransport", lambda cfg: transport)
@@ -143,8 +147,9 @@ class Operator:
         self.transport.receive(OPERATOR, text, thread=thread)
         self.settle()
 
-    def drive_until(self, wait: str, issue: int = 0) -> Any:
-        """Answer whatever the agent asks until it waits for ``wait``."""
+    def drive_until(self, wait: str, issue: int = 0, *, go: bool = True) -> Any:
+        """Answer whatever the agent asks until it waits for ``wait``.
+        Without ``go``, a restatement is left for its grace period."""
         deadline = time.monotonic() + DEADLINE_S
         answers = 0
         while time.monotonic() < deadline:
@@ -155,7 +160,13 @@ class Operator:
                 return item
             if item.status == "closed":
                 raise AssertionError(f"closed early: {self.last()}")
-            if item.position == WAITING and loop.wait == GO_AHEAD:
+            if wait == GO_AHEAD and loop.started:
+                raise AssertionError(f"went past the go-ahead: {self.last()}")
+            if item.position == WAITING and loop.wait == GO_AHEAD and not go:
+                assert loop.grace_until, f"no grace period: {self.last()}"
+                time.sleep(2)
+                self.tick()  # what the agent's timer would do
+            elif item.position == WAITING and loop.wait == GO_AHEAD:
                 self.say("go", issue)
             elif item.position == WAITING and loop.wait == ANSWER:
                 answers += 1
@@ -310,3 +321,107 @@ def test_an_issue_blocked_on_another_waits_for_it(
         assert f"\n+def vowels_{run}" not in diff
     finally:
         operator.tidy(issues, prs)
+
+
+def label(gitea: GiteaClient, repo: str, name: str) -> int:
+    """The repository's label ``name``, made if it's missing."""
+    for existing in gitea._request("GET", f"/repos/{repo}/labels"):
+        if existing["name"] == name:
+            return int(existing["id"])
+    made = gitea._request(
+        "POST", f"/repos/{repo}/labels", json={"name": name, "color": "#ee0701"}
+    )
+    return int(made["id"])
+
+
+def test_a_bug_goes_ahead_after_its_grace_period(
+    running: tuple[RunningLoop, InMemoryTransport, str],
+) -> None:
+    """M2.3: a `bug` issue is reproduced and restated, and work starts when
+    the grace period runs out, with no go from the operator."""
+    agent, transport, run = running
+    operator = Operator(agent, transport)
+    repo, base = agent.config.gitea_repo, agent.config.base_branch
+    as_operator = operator_gitea(agent.config.gitea_url)
+    issues: list[int] = []
+    prs: list[int] = []
+    try:
+        # A real bug on this run's base branch, for the agent to reproduce.
+        code = (
+            f"def last_word_{run}(text):\n"
+            '    """The last word of text, or "" if it has none."""\n'
+            "    return text.split()[-1]\n"
+        )
+        as_operator._request(
+            "POST",
+            f"/repos/{repo}/contents/sandbox/last_word_{run}.py",
+            json={
+                "branch": base,
+                "content": base64.b64encode(code.encode()).decode(),
+                "message": f"last_word_{run}, with a bug",
+            },
+        )
+        data = as_operator._request(
+            "POST",
+            f"/repos/{repo}/issues",
+            json={
+                "title": f"last_word_{run} crashes on empty text",
+                "body": f"`last_word_{run}('')` in sandbox/last_word_{run}.py "
+                'raises IndexError; it should return "" as its docstring says.',
+                "labels": [label(as_operator, repo, "bug")],
+                "assignees": [agent.account],
+            },
+        )
+        issues.append(int(data["number"]))
+        operator.tick()
+        assert loop_of(operator.item(issues[0])).labels == ("bug",)
+        item = operator.drive_until(MERGE, issues[0], go=False)
+        prs.append(loop_of(item).pr)
+        orchestrator = agent.worker.call(
+            lambda: agent.store.transcript(item.id, "orchestrator").read()
+        )
+        said = [getattr(e.event, "text", "") for e in orchestrator]
+        assert any(t.endswith("from the bug grace period") for t in said), said
+    finally:
+        operator.tidy(issues, prs)
+
+
+def test_a_plain_yes_is_a_go_ahead(
+    running: tuple[RunningLoop, InMemoryTransport, str],
+) -> None:
+    """M2.3: "yep, that's right" to a restatement is read as a go-ahead,
+    and the agent says so before acting."""
+    agent, transport, run = running
+    operator = Operator(agent, transport)
+    repo = agent.config.gitea_repo
+    as_operator = operator_gitea(agent.config.gitea_url)
+    issues: list[int] = []
+    prs: list[int] = []
+    try:
+        issues.append(as_operator.create_issue(repo, f"vowels_{run}", task(run)))
+        as_operator._request(
+            "PATCH",
+            f"/repos/{repo}/issues/{issues[0]}",
+            json={"assignees": [agent.account]},
+        )
+        operator.tick()
+        for _ in range(MAX_ANSWERS):
+            operator.drive_until(GO_AHEAD, issues[0])
+            if "?" not in operator.last():
+                break
+            # A restatement with a question in it can't be agreed to as it
+            # stands: answer first (the reader rightly says no to a yes).
+            operator.say(
+                f"Not yet: a new function vowels_{run}, not reusing anything "
+                "that exists. Put up a new restatement with that.",
+                issues[0],
+            )
+        operator.say("Yep, that's right, sounds good.", issues[0])
+        said = [m.text for m in transport.outbound]
+        assert any(t.startswith("Taking that as a go-ahead on rev-") for t in said)
+        assert loop_of(operator.item(issues[0])).started
+        pr = loop_of(operator.item(issues[0])).pr
+        if pr:
+            prs.append(pr)
+    finally:
+        operator.tidy(issues, prs)