diff --git a/docs/architecture/llm-client-integration.md b/docs/architecture/llm-client-integration.md
index d8b53a0..2608cdd 100644
--- a/docs/architecture/llm-client-integration.md
+++ b/docs/architecture/llm-client-integration.md
@@ -116,6 +116,8 @@ exercised in real use, and only then get a client subcommand.
 
 The currently shipped scope (per `scripts/atocore_client.py`):
 
+### Stable operations (shipped since the client was introduced)
+
 | Subcommand | Purpose | API endpoint(s) |
 |---|---|---|
 | `health` | service status, mount + source readiness | `GET /health` |
@@ -138,46 +140,46 @@ The currently shipped scope (per `scripts/atocore_client.py`):
 | `debug-context` | last context pack inspection | `GET /debug/context` |
 | `ingest-sources` | ingest configured source dirs | `POST /ingest/sources` |
 
-That covers everything in the "stable operations" set today:
-project lifecycle, ingestion, project-state curation, retrieval and
-context build, retrieval-quality audit, health and stats inspection.
+### Phase 9 reflection loop (shipped after migration safety work)
+
+These were explicitly deferred in earlier versions of this doc
+pending "exercised workflow". The constraint was real — premature
+API freeze would have made it harder to iterate on the ergonomics —
+but the deferral ran into a bootstrap problem: you can't exercise
+the workflow in real Claude Code sessions without a usable client
+surface to drive it from. The fix is to ship a minimal Phase 9
+surface now and treat it as stable-but-refinable: adding new
+optional parameters is fine, renaming subcommands is not.
+
+| Subcommand | Purpose | API endpoint(s) |
+|---|---|---|
+| `capture` | record one interaction round-trip | `POST /interactions` |
+| `extract` | run the rule-based extractor (preview or persist) | `POST /interactions/{id}/extract` |
+| `reinforce-interaction` | backfill reinforcement on an existing interaction | `POST /interactions/{id}/reinforce` |
+| `list-interactions` | paginated list with filters | `GET /interactions` |
+| `get-interaction` | fetch one interaction by id | `GET /interactions/{id}` |
+| `queue` | list the candidate review queue | `GET /memory?status=candidate` |
+| `promote` | move a candidate memory to active | `POST /memory/{id}/promote` |
+| `reject` | mark a candidate memory invalid | `POST /memory/{id}/reject` |
+
+All 8 Phase 9 subcommands have test coverage in
+`tests/test_atocore_client.py` via mocked `request()`, including
+an end-to-end test that drives the full capture → extract → queue
+→ promote/reject cycle through the client.
+
+### Coverage summary
+
+That covers everything in the "stable operations" set AND the
+full Phase 9 reflection loop: project lifecycle, ingestion,
+project-state curation, retrieval, context build,
+retrieval-quality audit, health and stats inspection, interaction
+capture, candidate extraction, candidate review queue.
 
 ## What's intentionally NOT in scope today
 
-Three families of operations are explicitly **deferred** until
-their workflows have been exercised in real use:
+Two families of operations remain deferred:
 
-### 1. Memory review queue and reflection loop
-
-Phase 9 Commit C shipped these endpoints:
-
-- `POST /interactions` (capture)
-- `POST /interactions/{id}/reinforce`
-- `POST /interactions/{id}/extract`
-- `GET /memory?status=candidate`
-- `POST /memory/{id}/promote`
-- `POST /memory/{id}/reject`
-
-The contracts are stable, but the **workflow ergonomics** are not.
-Until a real human has actually exercised the capture → extract →
-review → promote/reject loop a few times and we know what feels
-right, exposing those operations through the shared client would
-prematurely freeze a UX that's still being designed.
-
-When the loop has been exercised in real use and we know what
-the right subcommand shapes are, the shared client gains:
-
-- `capture <prompt> <response> [--project P] [--client C]`
-- `extract <interaction-id> [--persist]`
-- `queue` (list candidate review queue)
-- `promote <memory-id>`
-- `reject <memory-id>`
-
-At that point the Claude Code slash command can grow a companion
-`/atocore-record-response` command and the OpenClaw helper can be
-extended with the same flow.
-
-### 2. Backup and restore admin operations
+### 1. Backup and restore admin operations
 
 Phase 9 Commit B shipped these endpoints:
 
@@ -198,7 +200,7 @@ they would set `ATOCORE_FAIL_OPEN=false` for the duration of the
 call so the operator gets a real error on failure rather than a
 silent fail-open envelope.
 
-### 3. Engineering layer entity operations
+### 2. Engineering layer entity operations
 
 The engineering layer is in planning, not implementation. When
 V1 ships per `engineering-v1-acceptance.md`, the shared client
@@ -292,9 +294,15 @@ add a `CLIENT_VERSION = "x.y.z"` constant to
 
 1. **Refactor the OpenClaw helper** to shell out to the shared
    client. Cross-repo coordination, not blocking anything in
-   AtoCore itself.
-2. **Add memory-review subcommands** when the Phase 9 review
-   workflow has been exercised in real use.
+   AtoCore itself. With the Phase 9 subcommands now in the shared
+   client, the OpenClaw refactor can reuse all the reflection-loop
+   work instead of duplicating it.
+2. **Real-usage validation of the Phase 9 loop**, now that the
+   client surface exists. First capture → extract → review cycle
+   against the live Dalidou instance, likely via the Claude Code
+   slash command flow. Findings feed back into subcommand
+   refinement (new optional flags are fine, renames require a
+   semver bump).
 3. **Add backup admin subcommands** if and when we decide the
    shared client should be the canonical backup operator
    interface (with fail-open disabled for admin commands).
@@ -302,8 +310,9 @@ add a `CLIENT_VERSION = "x.y.z"` constant to
    engineering V1 implementation sprint, per
    `engineering-v1-acceptance.md`.
 5. **Tag a `CLIENT_VERSION` constant** the next time the shared
-   client surface meaningfully changes. Today's surface is the
-   v0.1.0 baseline.
+   client surface meaningfully changes. Today's surface with the
+   Phase 9 loop added is the v0.2.0 baseline (v0.1.0 was the
+   stable-ops-only version).
 
 ## TL;DR
 
@@ -314,9 +323,10 @@ add a `CLIENT_VERSION = "x.y.z"` constant to
   helper, future Codex skill, future MCP server) are thin
   wrappers that shell out to the shared client
 - The shared client today covers project lifecycle, ingestion,
-  retrieval, context build, project-state, and retrieval audit
-- Memory-review and engineering-entity commands are deferred
-  until their workflows are exercised
+  retrieval, context build, project-state, retrieval audit, AND
+  the full Phase 9 reflection loop (capture / extract /
+  reinforce / list / queue / promote / reject)
+- Backup admin and engineering-entity commands remain deferred
 - The OpenClaw helper is currently a parallel implementation and
   the refactor to the shared client is a queued follow-up
 - New LLM clients should never reimplement HTTP calls — they
diff --git a/scripts/atocore_client.py b/scripts/atocore_client.py
index ccfcff9..1592b28 100644
--- a/scripts/atocore_client.py
+++ b/scripts/atocore_client.py
@@ -23,6 +23,15 @@ TIMEOUT = int(os.environ.get("ATOCORE_TIMEOUT_SECONDS", "30"))
 REFRESH_TIMEOUT = int(os.environ.get("ATOCORE_REFRESH_TIMEOUT_SECONDS", "1800"))
 FAIL_OPEN = os.environ.get("ATOCORE_FAIL_OPEN", "true").lower() == "true"
 
+# Bumped when the subcommand surface or JSON output shapes meaningfully
+# change. See docs/architecture/llm-client-integration.md for the
+# semver rules. History:
+#   0.1.0  initial stable-ops-only client
+#   0.2.0  Phase 9 reflection loop added: capture, extract,
+#          reinforce-interaction, list-interactions, get-interaction,
+#          queue, promote, reject
+CLIENT_VERSION = "0.2.0"
+
 
 def print_json(payload: Any) -> None:
     print(json.dumps(payload, ensure_ascii=True, indent=2))
@@ -243,6 +252,59 @@ def build_parser() -> argparse.ArgumentParser:
     p.add_argument("top_k", nargs="?", type=int, default=5)
     p.add_argument("project", nargs="?", default="")
 
+    # --- Phase 9 reflection loop surface --------------------------------
+    #
+    # capture: record one interaction (prompt + response + context used).
+    #   Mirrors POST /interactions. response is positional so shell
+    #   callers can pass it via $(cat file.txt) or heredoc. project,
+    #   client, and session_id are optional positionals with empty
+    #   defaults, matching the existing script's style.
+    p = sub.add_parser("capture")
+    p.add_argument("prompt")
+    p.add_argument("response", nargs="?", default="")
+    p.add_argument("project", nargs="?", default="")
+    p.add_argument("client", nargs="?", default="")
+    p.add_argument("session_id", nargs="?", default="")
+    p.add_argument("reinforce", nargs="?", default="true")
+
+    # extract: run the Phase 9 C rule-based extractor against an
+    #   already-captured interaction. persist='true' writes the
+    #   candidates as status='candidate' memories; default is
+    #   preview-only.
+    p = sub.add_parser("extract")
+    p.add_argument("interaction_id")
+    p.add_argument("persist", nargs="?", default="false")
+
+    # reinforce: backfill reinforcement on an already-captured interaction.
+    p = sub.add_parser("reinforce-interaction")
+    p.add_argument("interaction_id")
+
+    # list-interactions: paginated listing with filters.
+    p = sub.add_parser("list-interactions")
+    p.add_argument("project", nargs="?", default="")
+    p.add_argument("session_id", nargs="?", default="")
+    p.add_argument("client", nargs="?", default="")
+    p.add_argument("since", nargs="?", default="")
+    p.add_argument("limit", nargs="?", type=int, default=50)
+
+    # get-interaction: fetch one by id
+    p = sub.add_parser("get-interaction")
+    p.add_argument("interaction_id")
+
+    # queue: list the candidate review queue
+    p = sub.add_parser("queue")
+    p.add_argument("memory_type", nargs="?", default="")
+    p.add_argument("project", nargs="?", default="")
+    p.add_argument("limit", nargs="?", type=int, default=50)
+
+    # promote: candidate -> active
+    p = sub.add_parser("promote")
+    p.add_argument("memory_id")
+
+    # reject: candidate -> invalid
+    p = sub.add_parser("reject")
+    p.add_argument("memory_id")
+
     return parser
 
 
@@ -304,6 +366,79 @@ def main() -> int:
         print_json(request("POST", "/context/build", {"prompt": args.prompt, "project": args.project or None, "budget": args.budget}))
     elif cmd == "audit-query":
         print_json(audit_query(args.prompt, args.top_k, args.project or None))
+    # --- Phase 9 reflection loop surface ------------------------------
+    elif cmd == "capture":
+        body: dict[str, Any] = {
+            "prompt": args.prompt,
+            "response": args.response,
+            "project": args.project,
+            "client": args.client or "atocore-client",
+            "session_id": args.session_id,
+            "reinforce": args.reinforce.lower() in {"1", "true", "yes", "y"},
+        }
+        print_json(request("POST", "/interactions", body))
+    elif cmd == "extract":
+        persist = args.persist.lower() in {"1", "true", "yes", "y"}
+        print_json(
+            request(
+                "POST",
+                f"/interactions/{urllib.parse.quote(args.interaction_id, safe='')}/extract",
+                {"persist": persist},
+            )
+        )
+    elif cmd == "reinforce-interaction":
+        print_json(
+            request(
+                "POST",
+                f"/interactions/{urllib.parse.quote(args.interaction_id, safe='')}/reinforce",
+                {},
+            )
+        )
+    elif cmd == "list-interactions":
+        query_parts: list[str] = []
+        if args.project:
+            query_parts.append(f"project={urllib.parse.quote(args.project)}")
+        if args.session_id:
+            query_parts.append(f"session_id={urllib.parse.quote(args.session_id)}")
+        if args.client:
+            query_parts.append(f"client={urllib.parse.quote(args.client)}")
+        if args.since:
+            query_parts.append(f"since={urllib.parse.quote(args.since)}")
+        query_parts.append(f"limit={int(args.limit)}")
+        query = "?" + "&".join(query_parts)
+        print_json(request("GET", f"/interactions{query}"))
+    elif cmd == "get-interaction":
+        print_json(
+            request(
+                "GET",
+                f"/interactions/{urllib.parse.quote(args.interaction_id, safe='')}",
+            )
+        )
+    elif cmd == "queue":
+        query_parts = ["status=candidate"]
+        if args.memory_type:
+            query_parts.append(f"memory_type={urllib.parse.quote(args.memory_type)}")
+        if args.project:
+            query_parts.append(f"project={urllib.parse.quote(args.project)}")
+        query_parts.append(f"limit={int(args.limit)}")
+        query = "?" + "&".join(query_parts)
+        print_json(request("GET", f"/memory{query}"))
+    elif cmd == "promote":
+        print_json(
+            request(
+                "POST",
+                f"/memory/{urllib.parse.quote(args.memory_id, safe='')}/promote",
+                {},
+            )
+        )
+    elif cmd == "reject":
+        print_json(
+            request(
+                "POST",
+                f"/memory/{urllib.parse.quote(args.memory_id, safe='')}/reject",
+                {},
+            )
+        )
     else:
         return 1
     return 0
diff --git a/tests/test_atocore_client.py b/tests/test_atocore_client.py
new file mode 100644
index 0000000..79b4227
--- /dev/null
+++ b/tests/test_atocore_client.py
@@ -0,0 +1,313 @@
+"""Tests for scripts/atocore_client.py — the shared operator CLI.
+
+Specifically covers the Phase 9 reflection-loop subcommands added
+after codex's sequence-step-3 review: ``capture``, ``extract``,
+``reinforce-interaction``, ``list-interactions``, ``get-interaction``,
+``queue``, ``promote``, ``reject``.
+
+The tests mock the client's ``request()`` helper and verify each
+subcommand:
+
+- calls the correct HTTP method and path
+- builds the correct JSON body (or the correct query string)
+- passes the right subset of CLI arguments through
+
+This is the same "wiring test" shape used by tests/test_api_storage.py:
+we don't exercise the live HTTP stack; we verify the client builds
+the request correctly. The server side is already covered by its
+own route tests.
+"""
+
+from __future__ import annotations
+
+import json
+import sys
+from pathlib import Path
+
+import pytest
+
+# Make scripts/ importable
+_REPO_ROOT = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(_REPO_ROOT / "scripts"))
+
+import atocore_client as client  # noqa: E402
+
+
+# ---------------------------------------------------------------------------
+# Request capture helper
+# ---------------------------------------------------------------------------
+
+
+class _RequestCapture:
+    """Drop-in replacement for client.request() that records calls."""
+
+    def __init__(self, response: dict | None = None):
+        self.calls: list[dict] = []
+        self._response = response if response is not None else {"ok": True}
+
+    def __call__(self, method, path, data=None, timeout=None):
+        self.calls.append(
+            {"method": method, "path": path, "data": data, "timeout": timeout}
+        )
+        return self._response
+
+
+@pytest.fixture
+def capture_requests(monkeypatch):
+    """Replace client.request with a recording stub and return it."""
+    stub = _RequestCapture()
+    monkeypatch.setattr(client, "request", stub)
+    return stub
+
+
+def _run_client(monkeypatch, argv: list[str]) -> int:
+    """Simulate a CLI invocation with the given argv."""
+    monkeypatch.setattr(sys, "argv", ["atocore_client.py", *argv])
+    return client.main()
+
+
+# ---------------------------------------------------------------------------
+# capture
+# ---------------------------------------------------------------------------
+
+
+def test_capture_posts_to_interactions_endpoint(capture_requests, monkeypatch):
+    _run_client(
+        monkeypatch,
+        [
+            "capture",
+            "what is p05's current focus",
+            "The current focus is wave 2 operational ingestion.",
+            "p05-interferometer",
+            "claude-code-test",
+            "session-abc",
+        ],
+    )
+    assert len(capture_requests.calls) == 1
+    call = capture_requests.calls[0]
+    assert call["method"] == "POST"
+    assert call["path"] == "/interactions"
+    body = call["data"]
+    assert body["prompt"] == "what is p05's current focus"
+    assert body["response"].startswith("The current focus")
+    assert body["project"] == "p05-interferometer"
+    assert body["client"] == "claude-code-test"
+    assert body["session_id"] == "session-abc"
+    assert body["reinforce"] is True  # default
+
+
+def test_capture_sets_default_client_when_omitted(capture_requests, monkeypatch):
+    _run_client(
+        monkeypatch,
+        ["capture", "hi", "hello"],
+    )
+    call = capture_requests.calls[0]
+    assert call["data"]["client"] == "atocore-client"
+    assert call["data"]["project"] == ""
+    assert call["data"]["reinforce"] is True
+
+
+def test_capture_accepts_reinforce_false(capture_requests, monkeypatch):
+    _run_client(
+        monkeypatch,
+        ["capture", "prompt", "response", "p05", "claude", "sess", "false"],
+    )
+    call = capture_requests.calls[0]
+    assert call["data"]["reinforce"] is False
+
+
+# ---------------------------------------------------------------------------
+# extract
+# ---------------------------------------------------------------------------
+
+
+def test_extract_default_is_preview(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["extract", "abc-123"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "POST"
+    assert call["path"] == "/interactions/abc-123/extract"
+    assert call["data"] == {"persist": False}
+
+
+def test_extract_persist_true(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["extract", "abc-123", "true"])
+    call = capture_requests.calls[0]
+    assert call["data"] == {"persist": True}
+
+
+def test_extract_url_encodes_interaction_id(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["extract", "abc/def"])
+    call = capture_requests.calls[0]
+    assert call["path"] == "/interactions/abc%2Fdef/extract"
+
+
+# ---------------------------------------------------------------------------
+# reinforce-interaction
+# ---------------------------------------------------------------------------
+
+
+def test_reinforce_interaction_posts_to_correct_path(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["reinforce-interaction", "int-xyz"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "POST"
+    assert call["path"] == "/interactions/int-xyz/reinforce"
+    assert call["data"] == {}
+
+
+# ---------------------------------------------------------------------------
+# list-interactions
+# ---------------------------------------------------------------------------
+
+
+def test_list_interactions_no_filters(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["list-interactions"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "GET"
+    assert call["path"] == "/interactions?limit=50"
+
+
+def test_list_interactions_with_project_filter(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["list-interactions", "p05-interferometer"])
+    call = capture_requests.calls[0]
+    assert "project=p05-interferometer" in call["path"]
+    assert "limit=50" in call["path"]
+
+
+def test_list_interactions_full_filter_set(capture_requests, monkeypatch):
+    _run_client(
+        monkeypatch,
+        [
+            "list-interactions",
+            "p05",
+            "sess-1",
+            "claude-code",
+            "2026-04-07T00:00:00Z",
+            "20",
+        ],
+    )
+    call = capture_requests.calls[0]
+    path = call["path"]
+    assert "project=p05" in path
+    assert "session_id=sess-1" in path
+    assert "client=claude-code" in path
+    # Since is URL-encoded — the : and + chars get escaped
+    assert "since=2026-04-07" in path
+    assert "limit=20" in path
+
+
+# ---------------------------------------------------------------------------
+# get-interaction
+# ---------------------------------------------------------------------------
+
+
+def test_get_interaction_fetches_by_id(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["get-interaction", "int-42"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "GET"
+    assert call["path"] == "/interactions/int-42"
+
+
+# ---------------------------------------------------------------------------
+# queue
+# ---------------------------------------------------------------------------
+
+
+def test_queue_always_filters_by_candidate_status(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["queue"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "GET"
+    assert call["path"].startswith("/memory?")
+    assert "status=candidate" in call["path"]
+    assert "limit=50" in call["path"]
+
+
+def test_queue_with_memory_type_and_project(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["queue", "adaptation", "p05-interferometer", "10"])
+    call = capture_requests.calls[0]
+    path = call["path"]
+    assert "status=candidate" in path
+    assert "memory_type=adaptation" in path
+    assert "project=p05-interferometer" in path
+    assert "limit=10" in path
+
+
+def test_queue_limit_coercion(capture_requests, monkeypatch):
+    """limit is typed as int by argparse so string '25' becomes 25."""
+    _run_client(monkeypatch, ["queue", "", "", "25"])
+    call = capture_requests.calls[0]
+    assert "limit=25" in call["path"]
+
+
+# ---------------------------------------------------------------------------
+# promote / reject
+# ---------------------------------------------------------------------------
+
+
+def test_promote_posts_to_memory_promote_path(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["promote", "mem-abc"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "POST"
+    assert call["path"] == "/memory/mem-abc/promote"
+    assert call["data"] == {}
+
+
+def test_reject_posts_to_memory_reject_path(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["reject", "mem-xyz"])
+    call = capture_requests.calls[0]
+    assert call["method"] == "POST"
+    assert call["path"] == "/memory/mem-xyz/reject"
+    assert call["data"] == {}
+
+
+def test_promote_url_encodes_memory_id(capture_requests, monkeypatch):
+    _run_client(monkeypatch, ["promote", "mem/with/slashes"])
+    call = capture_requests.calls[0]
+    assert "mem%2Fwith%2Fslashes" in call["path"]
+
+
+# ---------------------------------------------------------------------------
+# end-to-end: ensure the Phase 9 loop can be driven entirely through
+# the client
+# ---------------------------------------------------------------------------
+
+
+def test_phase9_full_loop_via_client_shape(capture_requests, monkeypatch):
+    """Simulate the full capture -> extract -> queue -> promote cycle.
+
+    This doesn't exercise real HTTP — each call is intercepted by
+    the mock request. But it proves every step of the Phase 9 loop
+    is reachable through the shared client, which is the whole point
+    of the codex-step-3 work.
+    """
+    # Step 1: capture
+    _run_client(
+        monkeypatch,
+        [
+            "capture",
+            "what about GF-PTFE for lateral support",
+            "## Decision: use GF-PTFE pads for thermal stability",
+            "p05-interferometer",
+        ],
+    )
+    # Step 2: extract candidates (preview)
+    _run_client(monkeypatch, ["extract", "fake-interaction-id"])
+    # Step 3: extract and persist
+    _run_client(monkeypatch, ["extract", "fake-interaction-id", "true"])
+    # Step 4: list the review queue
+    _run_client(monkeypatch, ["queue"])
+    # Step 5: promote a candidate
+    _run_client(monkeypatch, ["promote", "fake-memory-id"])
+    # Step 6: reject another
+    _run_client(monkeypatch, ["reject", "fake-memory-id-2"])
+
+    methods_and_paths = [
+        (c["method"], c["path"]) for c in capture_requests.calls
+    ]
+    assert methods_and_paths == [
+        ("POST", "/interactions"),
+        ("POST", "/interactions/fake-interaction-id/extract"),
+        ("POST", "/interactions/fake-interaction-id/extract"),
+        ("GET", "/memory?status=candidate&limit=50"),
+        ("POST", "/memory/fake-memory-id/promote"),
+        ("POST", "/memory/fake-memory-id-2/reject"),
+    ]