cloud ask: human-rendered output by default, JSON=1 to switch

Mirrors `make query` ergonomics:
  - default = pretty terminal layout (audit_label, sources w/ roles,
    bucket stats footer)
  - JSON=1 (or --json) = full machine-readable record

_render_cloud_ask_human shares the audit-label primitive
(_render_audit_label) with the local query renderer so the
HYBRID/STRICT/UNGROUNDED tokens map to the same four-rung ladder
labels in lattice modes (POINTER-LINKED / ANCHOR-WARRANTED / ...).
Bucket-direct path doesn't emit warrant-tail / run-DAG /
retrieval-purity so those sections are trimmed.

Output footer adds 'bucket: N HTTP requests · KB · endpoint / model'
so the operator can see network cost + LLM identity inline.

Makefile

@@ -1572,13 +1572,14 @@ cloud-fetch-chunk: bootstrap ## fetch + hash-verify one chunk from a bucket [LEA
 	@test -n "$(BLOB_BASE)" || { echo 'BLOB_BASE required (per-chunk blobs/<hash> base URL)'; exit 2; }
 	@$(ARBORIST) cloud fetch-chunk "$(LEAF_HASH)" --blob-base "$(BLOB_BASE)"
 
-cloud-ask: bootstrap ## bucket-direct end-to-end: FTS → LLM → verify [Q="..." SHARD_URL=... TOP_K=N MAX_CONTEXT=N CACHE_MB=N]
-	@test -n "$(Q)" || { echo 'usage: make cloud-ask Q="your question" [SHARD_URL=...] [TOP_K=4] [MAX_CONTEXT=24000] [CACHE_MB=64]'; exit 2; }
+cloud-ask: bootstrap ## bucket-direct end-to-end: FTS → LLM → verify [Q="..." JSON=1 SHARD_URL=... TOP_K=N MAX_CONTEXT=N CACHE_MB=N]
+	@test -n "$(Q)" || { echo 'usage: make cloud-ask Q="your question" [JSON=1] [SHARD_URL=...] [TOP_K=4] [MAX_CONTEXT=24000] [CACHE_MB=64]'; exit 2; }
 	@echo "# shard: $(SHARD_URL)" >&2
 	@$(ARBORIST) cloud ask '$(Q)' --shard-url "$(SHARD_URL)" \
 	    --top-k $(or $(TOP_K),4) \
 	    --max-context-chars $(or $(MAX_CONTEXT),24000) \
-	    --cache-mb $(or $(CACHE_MB),64)
+	    --cache-mb $(or $(CACHE_MB),64) \
+	    $(if $(JSON),--json,)
 
 # ---------------------------------------------------------------------------
 # Head-to-head: local query (BURN=1) vs cloud-ask, same question, same data.

arborist/cli.py

@@ -7290,6 +7290,11 @@ def build_parser() -> argparse.ArgumentParser:
         "--model", default=None,
         help="LLM model id (default: $ARBORIST_LLM_MODEL or Hermes-3-8B).",
     )
+    cloud_ask.add_argument(
+        "--json", action="store_true",
+        help="emit the full result as JSON. Default: human-rendered "
+        "(matches `arborist query` output).",
+    )
     cloud_ask.set_defaults(func=_cmd_cloud_ask)
 
     return p
@@ -7550,7 +7555,8 @@ def _cmd_cloud_ask(args: argparse.Namespace) -> int:
     finally:
         client.close()
 
-    print(json.dumps({
+    elapsed = _time.time() - t0
+    result = {
         "answer_text": answer,
         "audit_mode": verdict["audit_mode"],
         "verifier_method": verdict["verifier_method"],
@@ -7559,13 +7565,70 @@ def _cmd_cloud_ask(args: argparse.Namespace) -> int:
         "unverified_quotes": verdict.get("unverified_quotes", []),
         "sources": sources,
         "stats": client.stats(),
-        "timing_s": round(_time.time() - t0, 3),
+        "timing_s": round(elapsed, 3),
         "endpoint": endpoint,
         "model": model,
-    }, indent=2, ensure_ascii=False))
+    }
+    if getattr(args, "json", False):
+        print(json.dumps(result, indent=2, ensure_ascii=False))
+    else:
+        print(_render_cloud_ask_human(result, args.question))
     return 0
 
 
+def _render_cloud_ask_human(result: dict, question: str) -> str:
+    """Pretty-print a cloud-ask result. Mirrors `_render_query_human`'s
+    layout but trims sections that the bucket-direct path doesn't
+    produce (warrant tails, run-DAG, retrieval-purity)."""
+    audit = result.get("audit_mode") or "UNGROUNDED"
+    method = result.get("verifier_method") or "?"
+    n_quotes = result.get("n_quotes") or 0
+    n_verified = result.get("n_verified") or 0
+    elapsed_s = result.get("timing_s") or 0.0
+    label = _render_audit_label(audit, method, [])
+    stats = result.get("stats") or {}
+    cache_stats = stats.get("cache") or {}
+    http_reqs = stats.get("http_requests")
+    bytes_fetched = cache_stats.get("bytes_fetched") or 0
+
+    lines: list[str] = []
+    lines.append(question)
+    lines.append(
+        f"  {label}  {n_verified}/{n_quotes}  {elapsed_s:.2f}s  (bucket-direct)"
+    )
+    lines.append("")
+    lines.append(result.get("answer_text") or "")
+    lines.append("")
+
+    sources = result.get("sources") or []
+    if sources:
+        lines.append(f"sources ({len(sources)}):")
+        for i, s in enumerate(sources, start=1):
+            uri = s.get("document_uri", "")
+            title = (s.get("title") or "").strip() or _short_path(uri)
+            role = s.get("source_role")
+            role_part = f" — {role}" if role else ""
+            lines.append(
+                f"  [{i}] {title}{role_part} — {_strip_scheme(uri)}"
+            )
+        lines.append("")
+
+    unverified = result.get("unverified_quotes") or []
+    if unverified:
+        lines.append(f"unverified ({len(unverified)}):")
+        for q in unverified[:5]:
+            lines.append(f'  - "{q[:140]}{"…" if len(q) > 140 else ""}"')
+        lines.append("")
+
+    bytes_kb = bytes_fetched / 1024.0
+    lines.append(
+        f"bucket: {http_reqs} HTTP requests · {bytes_kb:.1f} KB · "
+        f"{result.get('endpoint','')} / {result.get('model','')}"
+    )
+    lines.append("  <run with --json for full record>")
+    return "\n".join(lines)
+
+
 def _cmd_cloud_fetch_chunk(args: argparse.Namespace) -> int:
     from arborist.merkle import hash_leaf
     client = _make_bucket_client(args)