ag2-memorysync 1.0.0__tar.gz

ag2_memorysync-1.0.0/.gitignore

@@ -0,0 +1,5 @@
+venv/
+dist/
+__pycache__/
+*.egg-info/
+.pytest_cache/

ag2_memorysync-1.0.0/PKG-INFO

@@ -0,0 +1,107 @@
+Metadata-Version: 2.5
+Name: ag2-memorysync
+Version: 1.0.0
+Summary: MemorySync for AG2 (AutoGen classic): an automatic memory loop on ConversableAgent's hook system — budgeted recall injection, duplicate-proof both-side capture, and a deadlock-free sync bridge.
+Project-URL: Homepage, https://docs.memorysync.io/guides/ag2
+Project-URL: Documentation, https://docs.memorysync.io/guides/ag2
+Project-URL: Repository, https://github.com/Rafay121/memorysync-plugins
+Author-email: MemorySync <support@memorysync.io>
+License-Expression: MIT
+Keywords: ag2,agents,autogen,conversableagent,long-term-memory,memory,memorysync
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Requires-Dist: httpx<1,>=0.25
+Description-Content-Type: text/markdown
+
+# ag2-memorysync
+
+[MemorySync](https://memorysync.io) for [AG2](https://github.com/ag2ai/ag2-classic)
+(the classic AutoGen `ConversableAgent` framework, `pip install autogen`):
+an automatic memory loop — recall injected before every reply, both sides
+persisted, zero extra code per turn.
+
+```bash
+pip install ag2-memorysync
+```
+
+## Quick start
+
+```python
+from autogen import ConversableAgent
+from ag2_memorysync import MemorySyncCapability
+
+assistant = ConversableAgent("assistant", llm_config=...)
+
+memory = MemorySyncCapability(
+    api_key="ms_...",              # or MEMORYSYNC_API_KEY
+    user_id="customer-42",         # required — who these memories belong to
+    session_id="support-chat",     # scopes the transcript
+)
+memory.add_to_agent(assistant)     # that's the whole integration
+```
+
+From then on every incoming user message is persisted and enriched with
+recalled context, and every outgoing reply is persisted — through AG2's
+own hook system (`process_last_received_message` +
+`process_message_before_send`).
+
+## Why this one
+
+| | Mem0 | Zep (`zep-ag2`) | **MemorySync** |
+| --- | --- | --- | --- |
+| AG2 adapter exists | ✗ docs show an AutoGen-0.2 recipe with placeholder model names | ✓ | ✓ |
+| Multi-agent duplication | — | ✗ **documented bug**: two attached agents store every utterance twice with conflicting roles | ✓ cross-hook dedup registry + idempotency seeds — one utterance, one row (by test) |
+| Sync→async bridge | — | per-call event-loop spin; documented deadlock caveat under asyncio | ✓ one persistent background loop; never touches the caller's loop (asyncio-driven chats pass, by test) |
+| Recall latency budget | — | ✗ none | ✓ hard 1.2s default — the reply is never late |
+| Framework pin | — | ✗ `ag2<1` — breaks on the v1 rewrite | ✓ **zero framework dependency** (duck-typed attach; works with whichever classic distribution you installed) |
+| Injected context re-stored? | — | system-message mutation, last-write-wins | ✓ hook output feeds the LLM only; the ORIGINAL text is what persists |
+
+## Semantics worth knowing
+
+- **The reply is never stalled and never broken.** Recall blocks at most
+  `recall_timeout` (default 1.2s); persistence is fire-and-forget off
+  the hot path. Outages and quota exhaustion degrade to "no memories
+  this turn".
+- Turns store verbatim under the `ag2::<session>` transcript scope with
+  deterministic idempotency seeds — retries and multi-agent echoes
+  converge on one stored row.
+- Tool/function messages are never persisted.
+- `register_memory_tools(memory, caller=..., executor=...)` adds
+  `search_memory` + `save_memory` tools (the caller needs an
+  `llm_config`, as usual for AG2 tools).
+- `memory.flush()` waits for in-flight writes (shutdown/tests);
+  `memory.close()` flushes and releases the HTTP client.
+
+## Configuration
+
+| Parameter | Default | Meaning |
+| --- | --- | --- |
+| `user_id` | — (required) | End user the memories belong to |
+| `session_id` | `"default"` | Transcript scope |
+| `top_k` | `5` | Memories considered per turn |
+| `recall_timeout` | `1.2` | Hard recall budget, seconds |
+| `min_prompt_chars` | `8` | Skip recall for trivial messages |
+| `context_template` | built-in | `{context}` placeholder, brace-safe `.replace` rendering |
+| `capture` | `"both"` | `"received"` / `"sent"` to capture one side only |
+
+## Development
+
+```bash
+pip install -e . "autogen[openai]" pytest
+python -m pytest tests -q      # 22 tests through REAL ConversableAgent chats
+```
+
+The suite includes a reproduction of zep-ag2's documented multi-agent
+double-store scenario (we store once) and a chat driven from inside
+`asyncio.run()` (no deadlock).
+
+## License
+
+MIT

ag2_memorysync-1.0.0/README.md

@@ -0,0 +1,85 @@
+# ag2-memorysync
+
+[MemorySync](https://memorysync.io) for [AG2](https://github.com/ag2ai/ag2-classic)
+(the classic AutoGen `ConversableAgent` framework, `pip install autogen`):
+an automatic memory loop — recall injected before every reply, both sides
+persisted, zero extra code per turn.
+
+```bash
+pip install ag2-memorysync
+```
+
+## Quick start
+
+```python
+from autogen import ConversableAgent
+from ag2_memorysync import MemorySyncCapability
+
+assistant = ConversableAgent("assistant", llm_config=...)
+
+memory = MemorySyncCapability(
+    api_key="ms_...",              # or MEMORYSYNC_API_KEY
+    user_id="customer-42",         # required — who these memories belong to
+    session_id="support-chat",     # scopes the transcript
+)
+memory.add_to_agent(assistant)     # that's the whole integration
+```
+
+From then on every incoming user message is persisted and enriched with
+recalled context, and every outgoing reply is persisted — through AG2's
+own hook system (`process_last_received_message` +
+`process_message_before_send`).
+
+## Why this one
+
+| | Mem0 | Zep (`zep-ag2`) | **MemorySync** |
+| --- | --- | --- | --- |
+| AG2 adapter exists | ✗ docs show an AutoGen-0.2 recipe with placeholder model names | ✓ | ✓ |
+| Multi-agent duplication | — | ✗ **documented bug**: two attached agents store every utterance twice with conflicting roles | ✓ cross-hook dedup registry + idempotency seeds — one utterance, one row (by test) |
+| Sync→async bridge | — | per-call event-loop spin; documented deadlock caveat under asyncio | ✓ one persistent background loop; never touches the caller's loop (asyncio-driven chats pass, by test) |
+| Recall latency budget | — | ✗ none | ✓ hard 1.2s default — the reply is never late |
+| Framework pin | — | ✗ `ag2<1` — breaks on the v1 rewrite | ✓ **zero framework dependency** (duck-typed attach; works with whichever classic distribution you installed) |
+| Injected context re-stored? | — | system-message mutation, last-write-wins | ✓ hook output feeds the LLM only; the ORIGINAL text is what persists |
+
+## Semantics worth knowing
+
+- **The reply is never stalled and never broken.** Recall blocks at most
+  `recall_timeout` (default 1.2s); persistence is fire-and-forget off
+  the hot path. Outages and quota exhaustion degrade to "no memories
+  this turn".
+- Turns store verbatim under the `ag2::<session>` transcript scope with
+  deterministic idempotency seeds — retries and multi-agent echoes
+  converge on one stored row.
+- Tool/function messages are never persisted.
+- `register_memory_tools(memory, caller=..., executor=...)` adds
+  `search_memory` + `save_memory` tools (the caller needs an
+  `llm_config`, as usual for AG2 tools).
+- `memory.flush()` waits for in-flight writes (shutdown/tests);
+  `memory.close()` flushes and releases the HTTP client.
+
+## Configuration
+
+| Parameter | Default | Meaning |
+| --- | --- | --- |
+| `user_id` | — (required) | End user the memories belong to |
+| `session_id` | `"default"` | Transcript scope |
+| `top_k` | `5` | Memories considered per turn |
+| `recall_timeout` | `1.2` | Hard recall budget, seconds |
+| `min_prompt_chars` | `8` | Skip recall for trivial messages |
+| `context_template` | built-in | `{context}` placeholder, brace-safe `.replace` rendering |
+| `capture` | `"both"` | `"received"` / `"sent"` to capture one side only |
+
+## Development
+
+```bash
+pip install -e . "autogen[openai]" pytest
+python -m pytest tests -q      # 22 tests through REAL ConversableAgent chats
+```
+
+The suite includes a reproduction of zep-ag2's documented multi-agent
+double-store scenario (we store once) and a chat driven from inside
+`asyncio.run()` (no deadlock).
+
+## License
+
+MIT

ag2_memorysync-1.0.0/pyproject.toml

@@ -0,0 +1,46 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "ag2-memorysync"
+dynamic = ["version"]
+description = "MemorySync for AG2 (AutoGen classic): an automatic memory loop on ConversableAgent's hook system — budgeted recall injection, duplicate-proof both-side capture, and a deadlock-free sync bridge."
+readme = "README.md"
+license = "MIT"
+requires-python = ">=3.10"
+authors = [{ name = "MemorySync", email = "support@memorysync.io" }]
+keywords = ["ag2", "autogen", "conversableagent", "agents", "memory", "memorysync", "long-term-memory"]
+classifiers = [
+    "Development Status :: 5 - Production/Stable",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+]
+# Deliberately NO dependency on any AutoGen distribution: the ecosystem
+# splits across `autogen` (ag2-classic), `pyautogen` (redirect), and
+# `ag2` (the v1 rewrite). The capability duck-types against the agent
+# instance it is attached to, so it works with whichever classic
+# distribution the application already installed — and can never create
+# a version conflict (Zep's zep-ag2 pins `ag2<1` and will break).
+dependencies = [
+    "httpx>=0.25,<1",
+]
+
+[project.urls]
+Homepage = "https://docs.memorysync.io/guides/ag2"
+Documentation = "https://docs.memorysync.io/guides/ag2"
+Repository = "https://github.com/Rafay121/memorysync-plugins"
+
+[tool.hatch.version]
+path = "src/ag2_memorysync/_version.py"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/ag2_memorysync"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]

ag2_memorysync-1.0.0/src/ag2_memorysync/__init__.py

@@ -0,0 +1,15 @@
+"""MemorySync for AG2 (AutoGen classic ConversableAgent framework)."""
+
+from ._api import MemorySyncAPIError, fnv1a64
+from ._version import __version__
+from .capability import DEFAULT_CONTEXT_TEMPLATE, MemorySyncCapability
+from .tools import register_memory_tools
+
+__all__ = [
+    "MemorySyncCapability",
+    "MemorySyncAPIError",
+    "DEFAULT_CONTEXT_TEMPLATE",
+    "register_memory_tools",
+    "fnv1a64",
+    "__version__",
+]

ag2_memorysync-1.0.0/src/ag2_memorysync/_api.py

@@ -0,0 +1,291 @@
+"""Async client for the MemorySync v1 data plane used by this adapter.
+
+Conversation turns persist through the *episodic* ingestion path
+(``POST /v1/memory/add_turn``), which stores text verbatim — no fact
+extraction, no low-value-chatter gate, no rewriting. An agent transcript
+must round-trip byte-for-byte; a plane that second-guessed it would
+corrupt the user's history.
+
+Everything here is async-native: it lives on the bridge's dedicated
+event loop, and AG2's synchronous hooks submit work to it without ever
+touching the caller's loop.
+"""
+
+from __future__ import annotations
+
+import os
+from typing import Any, Dict, List, Optional
+
+import httpx
+
+from ._version import __version__
+
+DEFAULT_BASE_URL = "https://api.memorysync.io"
+_USER_AGENT = f"ag2-memorysync/{__version__}"
+
+#: Namespace used when the key cannot list projects (see resolve_tenant_id).
+FALLBACK_TENANT = "default"
+
+#: One turn beyond this length is truncated before storage.
+MAX_TURN_CHARS = 16000
+
+
+class MemorySyncAPIError(Exception):
+    """A MemorySync call failed. Carries the status code and server detail."""
+
+    def __init__(self, message: str, *, status_code: Optional[int] = None) -> None:
+        super().__init__(message)
+        self.status_code = status_code
+
+
+def resolve_api_key(api_key: Optional[str]) -> str:
+    key = api_key or os.environ.get("MEMORYSYNC_API_KEY", "")
+    if not key or not key.strip():
+        raise ValueError(
+            "A MemorySync API key is required. Pass api_key=... or set the "
+            "MEMORYSYNC_API_KEY environment variable."
+        )
+    return key.strip()
+
+
+def resolve_base_url(base_url: Optional[str]) -> str:
+    url = base_url or os.environ.get("MEMORYSYNC_BASE_URL", "") or DEFAULT_BASE_URL
+    return url.rstrip("/")
+
+
+def fnv1a64(value: str) -> str:
+    """FNV-1a 64-bit over UTF-16 code units, as a fixed-width hex string.
+
+    Over UTF-16 code units — not code points, not UTF-8 bytes — so the
+    output matches every other MemorySync adapter (Python and JS)
+    character for character. Identical seeds across surfaces mean a turn
+    persisted here and again elsewhere converge on one stored row.
+    """
+    prime = 0x100000001B3
+    mask = 0xFFFFFFFFFFFFFFFF
+    h = 0xCBF29CE484222325
+    data = value.encode("utf-16-le")
+    for i in range(0, len(data), 2):
+        unit = data[i] | (data[i + 1] << 8)
+        h ^= unit
+        h = (h * prime) & mask
+    return format(h, "016x")
+
+
+class AsyncV1Api:
+    """Minimal asynchronous v1 client: add_turn, recall, query, list, forget."""
+
+    def __init__(
+        self,
+        *,
+        api_key: str,
+        base_url: str,
+        project_id: Optional[str] = None,
+        timeout: float = 30.0,
+        transport: Optional[httpx.AsyncBaseTransport] = None,
+    ) -> None:
+        self._api_key = api_key
+        self._base_url = base_url.rstrip("/")
+        self._project_id = project_id
+        self._timeout = timeout
+        self._transport = transport
+        self._http = httpx.AsyncClient(timeout=timeout, transport=transport)
+        self._tenant_id: Optional[str] = None
+        self._tenant_is_fallback = False
+
+    async def aclose(self) -> None:
+        await self._http.aclose()
+
+    @property
+    def base_url(self) -> str:
+        return self._base_url
+
+    # ── plumbing ─────────────────────────────────────────────────────
+
+    def _headers(self, *, end_user_id: Optional[str] = None) -> Dict[str, str]:
+        h = {
+            "X-API-Key": self._api_key,
+            "Accept": "application/json",
+            "User-Agent": _USER_AGENT,
+        }
+        if self._project_id:
+            h["X-Project-ID"] = self._project_id
+        if end_user_id:
+            h["X-End-User-ID"] = end_user_id
+        return h
+
+    async def _request(
+        self,
+        method: str,
+        path: str,
+        *,
+        json: Optional[Dict[str, Any]] = None,
+        params: Optional[Dict[str, Any]] = None,
+        end_user_id: Optional[str] = None,
+    ) -> Any:
+        url = f"{self._base_url}{path}"
+        try:
+            response = await self._http.request(
+                method,
+                url,
+                headers=self._headers(end_user_id=end_user_id),
+                json=json,
+                params=params,
+            )
+        except httpx.TimeoutException as e:
+            raise MemorySyncAPIError(f"Request timed out: {e}") from e
+        except httpx.HTTPError as e:
+            raise MemorySyncAPIError(f"Network error: {e}") from e
+
+        if response.status_code == 204:
+            return None
+        try:
+            body: Any = response.json()
+        except ValueError:
+            body = response.text or None
+        if response.status_code >= 400:
+            detail = body.get("detail") if isinstance(body, dict) else body
+            raise MemorySyncAPIError(
+                f"{method} {path} failed with HTTP {response.status_code}: {detail}",
+                status_code=response.status_code,
+            )
+        return body
+
+    # ── calls ────────────────────────────────────────────────────────
+
+    async def resolve_tenant_id(self) -> str:
+        """The tenant id, which the v1 routes need in path or body.
+
+        Derived from the project listing rather than asked for. Cached for
+        the lifetime of this client. Keys without the ``projects:read``
+        scope (evaluation keys) fall back to the fixed namespace
+        ``"default"`` — deterministic, so every read and write through
+        this client lands in one namespace. Only a definite 401/403
+        triggers the fallback; a transient server error re-raises rather
+        than silently switching namespaces.
+        """
+        if self._tenant_id:
+            return self._tenant_id
+        try:
+            projects = await self._request("GET", "/org/projects")
+        except MemorySyncAPIError as exc:
+            if exc.status_code in (401, 403):
+                self._tenant_id = FALLBACK_TENANT
+                self._tenant_is_fallback = True
+                return self._tenant_id
+            raise
+        first = projects[0] if isinstance(projects, list) and projects else None
+        tenant = first.get("tenant_id") if isinstance(first, dict) else None
+        if not tenant:
+            raise MemorySyncAPIError(
+                "Could not determine the tenant for this API key. Pass "
+                "tenant_id explicitly, or verify the key with `memorysync doctor`."
+            )
+        self._tenant_id = str(tenant)
+        return self._tenant_id
+
+    def set_tenant_id(self, tenant_id: str) -> None:
+        self._tenant_id = tenant_id
+
+    async def add_turn(
+        self,
+        *,
+        tenant_id: str,
+        user_id: str,
+        text: str,
+        speaker: Optional[str] = None,
+        metadata: Optional[Dict[str, Any]] = None,
+        source: str = "ag2",
+        sync_embed: bool = False,
+    ) -> Dict[str, Any]:
+        """Store one item verbatim (episodic ingestion).
+
+        ``speaker`` participates in the server's idempotency seed, so
+        retrying an identical payload is recognised
+        (``already_exists: true``) instead of stored twice.
+        """
+        body: Dict[str, Any] = {
+            "tenant_id": tenant_id,
+            "user_id": user_id,
+            "source": source,
+            "text": text,
+            "sync_embed": sync_embed,
+        }
+        if speaker is not None:
+            body["speaker"] = speaker
+        if metadata is not None:
+            body["metadata"] = metadata
+        return await self._request("POST", "/v1/memory/add_turn", json=body) or {}
+
+    async def recall(
+        self,
+        *,
+        tenant_id: str,
+        user_id: str,
+        prompt: str,
+        k: Optional[int] = None,
+    ) -> Dict[str, Any]:
+        """Hierarchical recall: grouped, prompt-ready context block."""
+        body: Dict[str, Any] = {
+            "tenant_id": tenant_id,
+            "user_id": user_id,
+            "prompt": prompt,
+        }
+        if k is not None:
+            body["k"] = k
+        return await self._request("POST", "/v1/memory/recall", json=body) or {}
+
+    async def query(
+        self,
+        *,
+        tenant_id: str,
+        user_id: str,
+        prompt: str,
+        k: Optional[int] = None,
+    ) -> Dict[str, Any]:
+        """Plain semantic search over the pair's memories (episodic included)."""
+        body: Dict[str, Any] = {
+            "tenant_id": tenant_id,
+            "user_id": user_id,
+            "prompt": prompt,
+        }
+        if k is not None:
+            body["k"] = k
+        return await self._request("POST", "/v1/memory/query", json=body) or {}
+
+    async def list_memories(
+        self,
+        *,
+        tenant_id: str,
+        user_id: str,
+        limit: int = 0,
+    ) -> List[Dict[str, Any]]:
+        """Every memory for the tenant/user pair, newest first."""
+        from urllib.parse import quote
+
+        raw = await self._request(
+            "GET",
+            f"/v1/memory/{quote(tenant_id, safe='')}/{quote(user_id, safe='')}/list",
+            params={"limit": limit},
+        )
+        memories = raw.get("memories") if isinstance(raw, dict) else None
+        return list(memories) if isinstance(memories, list) else []
+
+    async def add_memory(
+        self,
+        *,
+        user_id: str,
+        text: str,
+        source: str = "ag2",
+        metadata: Optional[Dict[str, Any]] = None,
+    ) -> Dict[str, Any]:
+        """Store a fact through the extraction path (server-side gating)."""
+        body: Dict[str, Any] = {"text": text, "source": source}
+        if metadata is not None:
+            body["metadata"] = metadata
+        return (
+            await self._request(
+                "POST", "/memory/add", json=body, end_user_id=user_id
+            )
+            or {}
+        )

ag2_memorysync-1.0.0/src/ag2_memorysync/_bridge.py

@@ -0,0 +1,96 @@
+"""One persistent background event loop bridging AG2's sync hooks to
+async MemorySync calls.
+
+AG2-classic's ``register_hook`` accepts only synchronous callables, so
+every memory integration must bridge to async I/O somehow. The naive
+bridge — spin up a fresh event loop per call — pays loop-startup latency
+on every turn and carries a documented deadlock caveat in Zep's adapter
+when the caller is already inside an async context.
+
+This bridge instead runs ONE daemon thread with ONE long-lived event
+loop per process. Callers submit coroutines with
+``run_coroutine_threadsafe``:
+
+- **The caller's own event loop is never touched.** Whether the hook
+  fires from plain ``initiate_chat`` or from inside somebody's
+  ``asyncio.run(...)``, the coroutine executes on the bridge's loop and
+  the hook thread blocks — bounded by an explicit timeout — on a plain
+  ``concurrent.futures`` future. No re-entrancy, no deadlock.
+- **Fire-and-forget writes stay off the hot path.** Persistence submits
+  and returns immediately; ``flush()`` awaits stragglers at shutdown or
+  in tests.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import concurrent.futures
+import logging
+import threading
+from typing import Any, Coroutine, Optional, Set
+
+logger = logging.getLogger("ag2_memorysync")
+
+
+class LoopBridge:
+    _shared: Optional["LoopBridge"] = None
+    _shared_lock = threading.Lock()
+
+    def __init__(self) -> None:
+        self._loop = asyncio.new_event_loop()
+        self._thread = threading.Thread(
+            target=self._run, name="memorysync-ag2-bridge", daemon=True
+        )
+        self._pending: Set[concurrent.futures.Future] = set()
+        self._pending_lock = threading.Lock()
+        self._thread.start()
+
+    @classmethod
+    def shared(cls) -> "LoopBridge":
+        with cls._shared_lock:
+            if cls._shared is None or not cls._shared._thread.is_alive():
+                cls._shared = cls()
+            return cls._shared
+
+    def _run(self) -> None:
+        asyncio.set_event_loop(self._loop)
+        self._loop.run_forever()
+
+    @property
+    def loop(self) -> asyncio.AbstractEventLoop:
+        return self._loop
+
+    def call(self, coro: Coroutine[Any, Any, Any], timeout: float) -> Any:
+        """Run a coroutine on the bridge loop; block at most ``timeout``.
+
+        On timeout the underlying task is cancelled so a slow backend
+        cannot pile up abandoned work.
+        """
+        future = asyncio.run_coroutine_threadsafe(coro, self._loop)
+        try:
+            return future.result(timeout)
+        except concurrent.futures.TimeoutError:
+            future.cancel()
+            raise
+
+    def submit(self, coro: Coroutine[Any, Any, Any], label: str) -> None:
+        """Fire-and-forget: schedule, log failures, never block."""
+        future = asyncio.run_coroutine_threadsafe(coro, self._loop)
+        with self._pending_lock:
+            self._pending.add(future)
+
+        def _done(f: concurrent.futures.Future) -> None:
+            with self._pending_lock:
+                self._pending.discard(f)
+            exc = f.exception() if not f.cancelled() else None
+            if exc is not None:
+                logger.warning("MemorySync %s skipped (%s)", label, exc)
+
+        future.add_done_callback(_done)
+
+    def flush(self, timeout: float = 10.0) -> None:
+        """Wait for all in-flight fire-and-forget work to land."""
+        with self._pending_lock:
+            pending = list(self._pending)
+        if pending:
+            concurrent.futures.wait(pending, timeout=timeout)

ag2_memorysync-1.0.0/src/ag2_memorysync/_version.py

@@ -0,0 +1 @@
+__version__ = "1.0.0"