A directory the model knows about, and @ to name a file in it

An agent chat used to open with the model knowing the name of a machine and
nothing about what was on it, so the first two rounds of every reply went on
finding out. It now gets a listing: one read-only command, `git ls-files` where
that works and `find` otherwise, falling back to an SFTP walk that always does.
git first because a repository already carries somebody's considered list of
what is not part of the project, and reproducing it by hand is how an index
ends up mostly build output.

The listing is budgeted rather than dumped. A tree of a thousand files is worse
than no tree -- it costs the window on every request forever and buries the four
names that mattered -- so directories that will not fit are shown as a count and
the model is told to open one itself. Collapsing picks the deepest and largest
first: by saving alone it would take `src/` before `src/web/static/vendor/`,
because it contains it, and lose every name worth having.

Read from a cache and never fetched. `harness.context_variables` is synchronous
and sits on the request path; the walk happens in the generation setup, which is
async and already doing network work, with a short wait. A chat whose first
reply outruns its first walk simply has no listing that turn and the fragment
disappears rather than appearing as an empty heading.

Then `@`, over the same index and over the library, and `/` for commands with an
Alt-based keyboard for the same jobs. A mentioned file arrives as contents, not
a reference -- a small model asked to call file_read often does not bother -- and
it arrives with its absolute path and the machine it came from, because a model
handed `main.py` cannot tell which of four it is and cannot name it back when
asked to change something.

The rule that matters for `/`: a message that merely starts with a slash still
sends. `//` escapes and an unrecognised command is posted as written. Swallowing
somebody's message is a much worse failure than an unknown command.

Two exceptions to Manual mode now, not one. Browsing and indexing are a person
acting, not a model, so neither passes through policy.py -- the same argument
the terminal panel rests on.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

src/lembas/api/chats.py

@@ -244,6 +244,53 @@ async def inspect_chat(request: Request, db: Db, user: RequiredUser, chat_id: st
     )
 
 
+@router.get("/{chat_id}/usage")
+async def chat_usage(request: Request, db: Db, user: RequiredUser, chat_id: str) -> Response:
+    """What this conversation has cost, and how full the window is.
+
+    Owner-checked and nothing else: it is your own chat's totals. Unlike the
+    inspector next door there is no admin branch, because there is no reason
+    for one -- the numbers describe a conversation, and reading somebody's
+    conversation is exactly what `sharing` has no admin branch for either.
+
+    Summed from what each reply recorded rather than recomputed: an endpoint
+    that reported no usage contributed an estimate at the time, and re-deriving
+    it now with a different estimator would make the totals move under a chat
+    that had not changed.
+    """
+    chat = _owned_chat(db, chat_id, user.id)
+    replies = list(
+        db.scalars(
+            select(Message)
+            .where(Message.chat_id == chat.id, Message.role == ROLE_ASSISTANT)
+            .order_by(Message.created_at)
+        )
+    )
+
+    totals = {"prompt": 0, "completion": 0, "total": 0}
+    estimated = False
+    for reply in replies:
+        usage = metrics_service.from_message(reply.usage_json)
+        totals["prompt"] += usage.prompt_tokens
+        totals["completion"] += usage.completion_tokens
+        totals["total"] += usage.total_tokens
+        estimated = estimated or usage.estimated
+
+    last = replies[-1] if replies else None
+    return render(
+        request,
+        "chat/_usage.html",
+        {
+            "chat": chat,
+            "totals": totals,
+            "estimated": estimated,
+            "replies": len(replies),
+            "metrics": metrics_service.from_message(last.usage_json if last else None),
+            "model": chat_service.model_for(db, chat),
+        },
+    )
+
+
 # Roughly what a downscaled phone photo comes to as base64. The exact figure
 # does not matter; putting megabytes of it into the DOM does.
 _REDACTED_URI = "data:…base64 image omitted…"