recall-langgraph 0.1.0__py3-none-any.whl

recall_langgraph/__init__.py

@@ -0,0 +1,20 @@
+"""Recall by Polign for LangGraph.
+
+``RecallResume`` resumes an agent from its own records: its working state,
+pointers to its work, relevant memories and recent turns come back as a
+briefing for the model, and the run's messages are recorded as turns for the
+next resume. ``recall_tools`` gives the model the tools that keep those
+records current. It sits beside your checkpointer; it does not replace it.
+"""
+
+from .resume import BRIEFING_ID, RecallResume
+from .tools import recall_tools
+
+__all__ = ["BRIEFING_ID", "RecallResume", "recall_tools"]
+
+try:
+    from importlib.metadata import version as _version
+
+    __version__ = _version("recall-langgraph")
+except Exception:  # pragma: no cover - source checkout without metadata
+    __version__ = "0.1.0"

recall_langgraph/resume.py

@@ -0,0 +1,347 @@
+"""One resumed agent, and the hooks that put it into a LangGraph graph."""
+
+from __future__ import annotations
+
+import asyncio
+import hashlib
+import json
+import os
+import threading
+from collections.abc import Mapping, Sequence
+from typing import Any
+
+from langchain_core.messages import (
+    AIMessage,
+    BaseMessage,
+    ChatMessage,
+    HumanMessage,
+    RemoveMessage,
+    SystemMessage,
+    ToolMessage,
+)
+from langchain_core.runnables import RunnableLambda
+from langgraph.graph.message import REMOVE_ALL_MESSAGES
+from polign_recall import Client, Output, Pointer, ResumeContext, ResumedAgent, WorkingState
+
+#: The id of the briefing message ``briefing_node`` adds to the state. The
+#: recorder skips it, so the briefing never becomes a turn of its own.
+BRIEFING_ID = "recall-briefing"
+
+
+class RecallResume:
+    """Resumes one agent from its Recall records and keeps them current.
+
+    Resuming takes the agent's lease and reads back the context to start
+    from: its working state, pointers to its work, relevant memories, and its
+    most recent turns, within ``token_budget``. ``briefing`` is that context
+    as one text for the model. The first use opens it; call ``open`` (or
+    ``aopen``) yourself to resume at a point you choose.
+
+    Old messages are not replayed into the state: the briefing already
+    carries the recent turns as text, and a replayed tool result without its
+    matching call would be refused by the model API. Your checkpointer keeps
+    working as usual. On a new thread id the resumed run starts from the
+    briefing alone; on the old thread id the checkpointer restores the old
+    messages too, and nothing is recorded twice: each turn keeps its message
+    id, so the first recording after a resume skips every restored message up
+    to the last one the records already have.
+
+    For ``create_react_agent``, pass ``pre_model_hook`` and
+    ``post_model_hook``. For your own StateGraph, call ``with_briefing`` on the
+    messages you send the model (or add ``briefing_node``), and add
+    ``record_node`` where new messages land.
+
+    ``client`` is a ``polign_recall.Client`` opened with ``agent=True``. When
+    it is left out, one is opened with ``local_dir`` and ``env`` and closed on
+    release. Release on a clean shutdown (``release``, ``arelease``, or a
+    ``with`` block) so the next process can resume at once instead of waiting
+    for the lease to expire.
+    """
+
+    def __init__(
+        self,
+        agent_id: str,
+        *,
+        client: Client | None = None,
+        token_budget: int | None = None,
+        lease_ttl: float | None = None,
+        holder: str | None = None,
+        output_threshold: int | None = None,
+        local_dir: str | os.PathLike[str] | None = None,
+        env: Mapping[str, str] | None = None,
+    ) -> None:
+        if client is not None and (local_dir is not None or env is not None):
+            raise ValueError("local_dir and env open a client; do not pass them with client")
+        self.agent_id = agent_id
+        self._client = client
+        self._owns_client = client is None
+        self._client_options: dict[str, Any] = {"local_dir": local_dir, "env": env}
+        self._resume_options: dict[str, Any] = {
+            "token_budget": token_budget,
+            "lease_ttl": lease_ttl,
+            "holder": holder,
+            "output_threshold": output_threshold,
+        }
+        self._agent: ResumedAgent | None = None
+        self._released = False
+        self._open_lock = threading.Lock()
+        # Serializes recording, so the pre and post hooks of parallel branches
+        # cannot record the same message twice.
+        self._record_lock = threading.Lock()
+        self._recorded: set[str] = set()
+        self._seeded = False
+
+    # Lifecycle.
+
+    def open(self) -> RecallResume:
+        """Resume the agent, once. Raises ``polign_recall.RecallError`` with
+        code "lease_held" while another process holds it."""
+        with self._open_lock:
+            if self._released:
+                raise RuntimeError("this resume was released; create a new one to resume again")
+            if self._agent is not None:
+                return self
+            if self._client is None:
+                self._client = Client(agent=True, **self._client_options)
+            try:
+                self._agent = self._client.resume(self.agent_id, **self._resume_options)
+            except BaseException:
+                if self._owns_client:
+                    self._client.close()
+                    self._client = None
+                raise
+            return self
+
+    async def aopen(self) -> RecallResume:
+        return await asyncio.to_thread(self.open)
+
+    def release(self) -> None:
+        """Hand the lease over and, if this object opened its own client,
+        close it. Safe to call more than once."""
+        with self._open_lock:
+            if self._released:
+                return
+            self._released = True
+            agent, client = self._agent, self._client
+            self._agent = None
+        try:
+            if agent is not None:
+                agent.release()
+        finally:
+            if self._owns_client and client is not None:
+                client.close()
+
+    async def arelease(self) -> None:
+        await asyncio.to_thread(self.release)
+
+    def __enter__(self) -> RecallResume:
+        return self.open()
+
+    def __exit__(self, *exc: object) -> None:
+        self.release()
+
+    async def __aenter__(self) -> RecallResume:
+        return await self.aopen()
+
+    async def __aexit__(self, *exc: object) -> None:
+        await self.arelease()
+
+    @property
+    def agent(self) -> ResumedAgent:
+        """The resumed agent, for the calls this class does not wrap. Opens it
+        on first use."""
+        if self._agent is None:
+            self.open()
+        assert self._agent is not None
+        return self._agent
+
+    @property
+    def context(self) -> ResumeContext:
+        """What the resume returned."""
+        return self.agent.context
+
+    @property
+    def briefing(self) -> str:
+        return self.context.briefing
+
+    # The briefing.
+
+    def briefing_message(self) -> SystemMessage:
+        return SystemMessage(content=self.briefing, id=BRIEFING_ID)
+
+    def with_briefing(self, messages: Sequence[BaseMessage]) -> list[BaseMessage]:
+        """``messages`` with the briefing in front, for the model's input. It
+        does not change the graph state. A briefing already there is not
+        added twice."""
+        messages = list(messages)
+        if any(m.id == BRIEFING_ID for m in messages):
+            return messages
+        return [self.briefing_message(), *messages]
+
+    def _pre(self, state: Any) -> dict[str, Any]:
+        messages = _messages(state)
+        self.record(messages)
+        return {"llm_input_messages": self.with_briefing(messages)}
+
+    async def _apre(self, state: Any) -> dict[str, Any]:
+        return await asyncio.to_thread(self._pre, state)
+
+    def _post(self, state: Any) -> dict[str, Any]:
+        self.record(_messages(state))
+        return {}
+
+    async def _apost(self, state: Any) -> dict[str, Any]:
+        return await asyncio.to_thread(self._post, state)
+
+    def _briefing_update(self, state: Any) -> dict[str, Any]:
+        messages = _messages(state)
+        if any(m.id == BRIEFING_ID for m in messages):
+            return {}
+        # add_messages appends, so putting the briefing first means rewriting
+        # the list, the way LangGraph documents for a pre-model hook.
+        return {"messages": [RemoveMessage(id=REMOVE_ALL_MESSAGES), self.briefing_message(), *messages]}
+
+    async def _abriefing_update(self, state: Any) -> dict[str, Any]:
+        return await asyncio.to_thread(self._briefing_update, state)
+
+    @property
+    def pre_model_hook(self) -> RunnableLambda:
+        """For ``create_react_agent(pre_model_hook=...)``. Records the messages
+        that arrived since the last model call and hands the model the
+        briefing followed by the state's messages, as ``llm_input_messages``,
+        so the state itself is unchanged. To combine it with a hook of your
+        own (trimming, say), call ``record`` and ``with_briefing`` from it."""
+        return RunnableLambda(self._pre, afunc=self._apre, name="recall_pre_model_hook")
+
+    @property
+    def post_model_hook(self) -> RunnableLambda:
+        """For ``create_react_agent(post_model_hook=...)``. Records the model's
+        reply as soon as it arrives, so a crash during the tool calls that
+        follow does not lose it."""
+        return RunnableLambda(self._post, afunc=self._apost, name="recall_post_model_hook")
+
+    @property
+    def record_node(self) -> RunnableLambda:
+        """A node for your own StateGraph that records every message not
+        recorded yet and changes nothing in the state."""
+        return RunnableLambda(self._post, afunc=self._apost, name="recall_record")
+
+    @property
+    def briefing_node(self) -> RunnableLambda:
+        """A node for your own StateGraph that adds the briefing to the
+        ``messages`` state as a SystemMessage, once per thread. Put it first.
+        Prefer ``with_briefing`` in your model node when you can: it keeps the
+        briefing out of the state."""
+        return RunnableLambda(self._briefing_update, afunc=self._abriefing_update, name="recall_briefing")
+
+    # Recording.
+
+    def record(self, messages: Sequence[BaseMessage]) -> int:
+        """Record each message not recorded yet, in order, as a turn. Returns
+        how many were recorded. Messages are told apart by id; the briefing
+        and messages without text are skipped."""
+        count = 0
+        with self._record_lock:
+            if not self._seeded:
+                self._seed(messages)
+                self._seeded = True
+            for message in messages:
+                key = _key(message)
+                if key in self._recorded or message.id == BRIEFING_ID:
+                    continue
+                turn = _turn(message)
+                if turn is not None:
+                    self.agent.record_turn(*turn, message_id=message.id or None)
+                    count += 1
+                self._recorded.add(key)
+        return count
+
+    def _seed(self, messages: Sequence[BaseMessage]) -> None:
+        """Mark messages a checkpointer restored as already recorded.
+
+        The previous process recorded in order, so every message up to the
+        last one whose id the records hold was recorded, whether or not its
+        own id made it into the recent turns. Anything after that is new: the
+        caller's fresh input, or a message the previous process got but died
+        before recording, and both should be recorded now."""
+        if self.agent.context.turn_seq == 0:
+            return
+        known = {t.message_id for t in self.agent.recent_turns(200) if t.message_id}
+        last = -1
+        for i, message in enumerate(messages):
+            if message.id and message.id in known:
+                last = i
+        for message in messages[: last + 1]:
+            self._recorded.add(_key(message))
+
+    async def arecord(self, messages: Sequence[BaseMessage]) -> int:
+        return await asyncio.to_thread(self.record, messages)
+
+    # What the model tools call.
+
+    def update_working_state(self, **fields: Any) -> WorkingState:
+        return self.agent.update_working_state(**fields)
+
+    def milestone(self, name: str, progress: str | None = None) -> WorkingState:
+        return self.agent.milestone(name, progress)
+
+    def fetch_output(self, ref: str) -> Output:
+        return self.agent.fetch_output(ref)
+
+    def set_pointer(self, name: str, type: str, fields: Mapping[str, str], note: str | None = None) -> Pointer:
+        return self.agent.set_pointer(name, type, fields, note)
+
+
+def _messages(state: Any) -> list[BaseMessage]:
+    messages = state.get("messages") if isinstance(state, Mapping) else getattr(state, "messages", None)
+    if messages is None:
+        raise ValueError("the graph state has no messages")
+    return list(messages)
+
+
+def _key(message: BaseMessage) -> str:
+    if message.id:
+        return message.id
+    # LangGraph's add_messages gives every message an id, so this is only for
+    # lists built by hand.
+    body = json.dumps([message.type, message.content, getattr(message, "tool_call_id", None)],
+                      default=str, sort_keys=True)
+    return "sha256:" + hashlib.sha256(body.encode()).hexdigest()
+
+
+def _text(content: Any) -> str:
+    """The readable text of a message: a string, or a list of content blocks."""
+    if isinstance(content, str):
+        return content
+    if not isinstance(content, list):
+        return ""
+    parts = []
+    for block in content:
+        if isinstance(block, str):
+            parts.append(block)
+        elif isinstance(block, dict):
+            if isinstance(block.get("text"), str):
+                parts.append(block["text"])
+            elif block.get("type") not in (None, "tool_use", "tool_call"):
+                parts.append(f"[{block['type']}]")
+    return "\n".join(p for p in parts if p)
+
+
+def _turn(message: BaseMessage) -> tuple[str, str, str | None] | None:
+    """The (role, content, name) turn one message is recorded as."""
+    text = _text(message.content)
+    if isinstance(message, HumanMessage):
+        return ("user", text, None) if text else None
+    if isinstance(message, AIMessage):
+        calls = [{"name": c["name"], "args": c["args"], "id": c.get("id")} for c in message.tool_calls]
+        if calls:
+            rendered = json.dumps({"tool_calls": calls}, ensure_ascii=False, default=str)
+            text = f"{text}\n{rendered}" if text else rendered
+        return ("assistant", text, None) if text else None
+    if isinstance(message, ToolMessage):
+        return "tool", text or json.dumps(message.content, default=str), message.name
+    if isinstance(message, SystemMessage):
+        return ("system", text, None) if text else None
+    if isinstance(message, ChatMessage) and message.role in ("system", "user", "assistant", "tool"):
+        return (message.role, text, None) if text else None
+    return ("assistant", text, None) if text else None

recall_langgraph/tools.py

@@ -0,0 +1,140 @@
+"""LangChain tools that let the model write what its next instance resumes from."""
+
+from __future__ import annotations
+
+import asyncio
+from collections.abc import Callable
+from typing import Any, Literal, TypeVar
+
+from langchain_core.tools import BaseTool, StructuredTool, ToolException
+from polign_recall import RecallError
+
+from .resume import RecallResume
+
+T = TypeVar("T")
+
+PointerType = Literal["git_ref", "object", "env", "external", "process"]
+
+
+def _call(fn: Callable[[], T]) -> T:
+    try:
+        return fn()
+    except RecallError as exc:
+        # The model sees this text as the tool's result and can react to it.
+        raise ToolException(str(exc)) from exc
+
+
+def _tool(fn: Callable[..., str]) -> BaseTool:
+    async def coroutine(**kwargs: Any) -> str:
+        return await asyncio.to_thread(fn, **kwargs)
+
+    return StructuredTool.from_function(
+        func=fn, coroutine=coroutine, parse_docstring=True, handle_tool_error=True
+    )
+
+
+def recall_tools(resume: RecallResume) -> list[BaseTool]:
+    """The tools to give the model: ``update_working_state``, ``milestone``,
+    ``fetch_output`` and ``set_pointer``. Each works with ``invoke`` and
+    ``ainvoke``. A failed call comes back to the model as the tool's text."""
+
+    def update_working_state(
+        goal: str | None = None,
+        plan: list[str] | None = None,
+        progress: str | None = None,
+        focus: str | None = None,
+        decisions: list[str] | None = None,
+        open_questions: list[str] | None = None,
+        notes: str | None = None,
+    ) -> str:
+        """Save your working state: the note your next instance resumes from if this run stops.
+
+        Fields you give replace the current ones; fields you leave out are kept. Write it at
+        natural boundaries, as a colleague taking over would need it.
+
+        Args:
+            goal: What the whole task is for.
+            plan: The remaining steps, in order.
+            progress: What is done so far.
+            focus: What you are doing right now.
+            decisions: Decisions made, each with its reason.
+            open_questions: What is still unknown.
+            notes: Anything else the next instance needs.
+        """
+        given = {
+            "goal": goal, "plan": plan, "progress": progress, "focus": focus,
+            "decisions": decisions, "open_questions": open_questions, "notes": notes,
+        }
+        state = _call(lambda: resume.update_working_state(**{k: v for k, v in given.items() if v is not None}))
+        return f"Working state saved (version {state.version})."
+
+    def milestone(name: str, progress: str | None = None) -> str:
+        """Declare that you reached a durable point, such as tests passing or a section drafted.
+
+        If this run stops, the next instance loses at most the work since the last milestone.
+
+        Args:
+            name: The milestone reached.
+            progress: What is now done.
+        """
+        _call(lambda: resume.milestone(name, progress))
+        return f"Milestone recorded: {name}."
+
+    def fetch_output(ref: str) -> str:
+        """Return a stored output in full, by the ref shown in your context.
+
+        Args:
+            ref: The output's ref.
+        """
+        return _call(lambda: resume.fetch_output(ref)).content
+
+    def set_pointer(
+        name: str,
+        type: PointerType,
+        repo: str | None = None,
+        branch: str | None = None,
+        sha: str | None = None,
+        uri: str | None = None,
+        image: str | None = None,
+        lockfile_hash: str | None = None,
+        setup: str | None = None,
+        kind: str | None = None,
+        id: str | None = None,
+        url: str | None = None,
+        command: str | None = None,
+        port: str | None = None,
+        note: str | None = None,
+    ) -> str:
+        """Record where a piece of your work lives, so your next instance can find it.
+
+        Replaces any pointer with the same name. Each type needs one of its fields:
+        git_ref (repo, branch, sha; needs sha or branch), object (uri), env (image,
+        lockfile_hash, setup), external (kind, id, url; needs id or url), process
+        (command, port; needs command).
+
+        Args:
+            name: A stable name for this piece of work.
+            type: Where it lives.
+            repo: git_ref: the repository.
+            branch: git_ref: the branch.
+            sha: git_ref: the commit.
+            uri: object: where the object is stored.
+            image: env: the container image.
+            lockfile_hash: env: the hash of the dependency lockfile.
+            setup: env: how to set the environment up.
+            kind: external: what kind of thing it is, such as a ticket.
+            id: external: its id.
+            url: external: its URL.
+            command: process: the command that runs it.
+            port: process: the port it listens on.
+            note: What it is.
+        """
+        given = {
+            "repo": repo, "branch": branch, "sha": sha, "uri": uri, "image": image,
+            "lockfile_hash": lockfile_hash, "setup": setup, "kind": kind, "id": id,
+            "url": url, "command": command, "port": port,
+        }
+        pointer = _call(lambda: resume.set_pointer(name, type, {k: v for k, v in given.items() if v}, note))
+        return f"Pointer {pointer.name} saved."
+
+    return [_tool(f) for f in (update_working_state, milestone, fetch_output, set_pointer)]

recall_langgraph-0.1.0.dist-info/METADATA

@@ -0,0 +1,209 @@
+Metadata-Version: 2.5
+Name: recall-langgraph
+Version: 0.1.0
+Summary: Resume LangGraph agents from their own records, backed by Recall by Polign
+Project-URL: Homepage, https://polign.com
+Project-URL: Documentation, https://polign.com/integrations.html
+Project-URL: Repository, https://github.com/Polign/polign
+Project-URL: Issues, https://github.com/Polign/polign/issues
+Author: Polign
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: agent memory,langchain,langgraph,polign,recall,resume
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Requires-Dist: langchain-core<2,>=1.0
+Requires-Dist: langgraph<2,>=1.0
+Requires-Dist: polign-recall>=0.4.0
+Description-Content-Type: text/markdown
+
+# recall-langgraph
+
+Resume [LangGraph](https://github.com/langchain-ai/langgraph) agents from
+their own records, backed by [Recall by Polign](https://polign.com/recall.html).
+
+A long-running agent that crashes, gets rescheduled, or moves to another
+machine usually starts over, or restores an exact snapshot of its state. With
+this package it picks up where it left off from what it wrote down instead:
+while it works, it keeps a working state (its goal, plan, progress and
+decisions) and pointers to where its work lives, and every message of the run
+is recorded. The next process that opens the same agent id gets a fresh
+starting context built from those records, sized to a token budget.
+
+This sits beside your checkpointer, not in place of it. LangGraph's
+checkpointer still handles threads and interrupts however you configure it.
+
+```bash
+pip install recall-langgraph
+```
+
+pip also brings the [`polign_db`](https://pypi.org/project/polign-db/)
+package with the `polign` CLI and `polign-server` binaries for Linux, macOS
+and Windows, so there is no separate database to download.
+
+## Usage with create_react_agent
+
+```python
+from langchain_openai import ChatOpenAI
+from langgraph.prebuilt import create_react_agent
+from recall_langgraph import RecallResume, recall_tools
+
+with RecallResume("billing-migrator", local_dir="./recall-data") as resume:
+    agent = create_react_agent(
+        ChatOpenAI(model="gpt-4.1"),
+        tools=[*your_tools, *recall_tools(resume)],
+        prompt=(
+            "You move the billing service to the v2 API. Keep your working state "
+            "current with update_working_state, and call milestone when a step is done."
+        ),
+        pre_model_hook=resume.pre_model_hook,
+        post_model_hook=resume.post_model_hook,
+    )
+    agent.invoke({"messages": [("user", "Carry on with the migration.")]})
+```
+
+Run it, stop it halfway, and run it again: the second run starts from the
+working state, pointers and recent turns the first one wrote.
+
+What happens:
+
+1. Entering the `with` block resumes the agent. It takes a lease on the id,
+   so only one process acts for the agent at a time, and reads back its
+   records. The first time an id is used, it starts fresh.
+2. Before every model call, `pre_model_hook` puts one `SystemMessage` in
+   front of the messages, the briefing: the working state, pointers to the
+   work, relevant memories, and the most recent turns, within
+   `token_budget`. It goes to the model as `llm_input_messages`, so the
+   graph's `messages` state never holds it.
+3. The hooks record each new message once, in order, as a turn: human
+   messages before the model call, the model's reply right after it, and tool
+   results before the next call. A reply with tool calls is recorded with the
+   calls as JSON text. A turn longer than the output threshold is stored whole
+   as an output, and the briefing shows a reference the model can pass to
+   `fetch_output`.
+4. The model keeps its working state current through the tools from
+   `recall_tools`.
+5. Leaving the block (or calling `release()`) hands the lease over, so the
+   next process can resume at once. A process that dies without releasing
+   holds the lease until it expires (60 seconds by default).
+
+Recall never replays old messages into the state: the briefing already
+carries the recent turns as text, and a replayed tool result without its
+matching tool call would be refused by the model API. On a new `thread_id`,
+the resumed run starts from the briefing alone. On the old `thread_id`, a
+persistent checkpointer restores the old messages as well, and nothing is
+recorded twice: each turn keeps its message id, so the first recording after
+a resume skips every restored message up to the last one Recall already has.
+Messages after that one, such as the caller's new input or a reply the dead
+process never recorded, are recorded as usual. The model then sees the
+restored history and the briefing both; start a new thread if you would
+rather it saw only the briefing.
+
+`create_react_agent` is deprecated in LangGraph 1.0 in favor of
+`langchain.agents.create_agent`, but it still ships in `langgraph.prebuilt`
+and the hooks work with it. Use `version="v2"` (the default), since
+`post_model_hook` needs it.
+
+## Your own StateGraph
+
+```python
+from langgraph.graph import START, MessagesState, StateGraph
+
+def call_model(state: MessagesState):
+    return {"messages": [model.invoke(resume.with_briefing(state["messages"]))]}
+
+builder = StateGraph(MessagesState)
+builder.add_node("model", call_model)
+builder.add_node("record", resume.record_node)
+builder.add_edge(START, "model")
+builder.add_edge("model", "record")
+```
+
+- `with_briefing(messages)` returns the messages with the briefing in front,
+  for the model's input, and leaves the state alone.
+- `record_node` records every message not recorded yet and changes nothing.
+  Put it after each node that adds messages. `record(messages)` does the same
+  from your own code.
+- `briefing_node` adds the briefing to the `messages` state itself, once per
+  thread, if you would rather keep it there. Put it first.
+
+The hooks and nodes work with `invoke` and `ainvoke`.
+
+## The tools
+
+| Tool | What the model uses it for |
+|---|---|
+| `update_working_state` | Save its goal, plan, progress, focus, decisions, open questions and notes. Fields it leaves out are kept. |
+| `milestone` | Declare a durable point, such as tests passing. |
+| `fetch_output` | Read a stored output in full by its ref. |
+| `set_pointer` | Record where a piece of work lives: a git branch or commit, an object, an environment, an external item, or a process. |
+
+A failed call comes back to the model as the tool's text. Your own code can
+call `resume.update_working_state`, `resume.milestone`, `resume.fetch_output`
+and `resume.set_pointer`, and `resume.agent` is the underlying
+`polign_recall.ResumedAgent` for everything else (`record_turn`,
+`recent_turns`, `store_output`, `pointers`, `working_state_history`).
+`resume.context` is what the resume returned, and `resume.briefing` its text.
+
+## Options
+
+| Option | Default | What it does |
+|---|---|---|
+| `client` | none | A `polign_recall.Client` opened with `agent=True`, to share one subprocess. Without it, `RecallResume` opens its own and closes it on release. |
+| `local_dir` | none | Keep the database in this directory and run its server (only when `RecallResume` opens its own client) |
+| `env` | process environment | `POLIGN_URL`, `POLIGN_API_KEY` and the rest, for a server you run yourself |
+| `token_budget` | 8000 | Bound on the briefing, in tokens |
+| `lease_ttl` | 60 | Seconds each lease epoch lasts; renewed in the background |
+| `holder` | host, pid and a random suffix | Names this process in lease records |
+| `output_threshold` | 2000 | Turns longer than this many tokens are stored as outputs |
+
+## Errors
+
+If another process holds the agent, resuming raises
+`polign_recall.RecallError` with code `"lease_held"`. Wait and try again, or
+stop. If another process takes the agent over mid-run, the next record fails
+with code `"lease_lost"`, which stops the graph run; this process should
+stop acting for the agent.
+
+Records are append-only. Removing or trimming messages in the graph state
+changes what the model sees, never what was recorded.
+
+## Where the records are stored
+
+`local_dir="./recall-data"` keeps the database on this machine; the first
+process to open the directory starts a `polign-server` for it in the
+background, and later ones share it. Agents that move between machines need
+one shared server: run `polign-server` where they can all reach it
+([Get started](https://polign.com/developers.html) shows how, including
+storing into S3, GCS or Azure) and pass its address through `env` or the
+environment:
+
+```python
+RecallResume(
+    "billing-migrator",
+    env={"POLIGN_URL": "http://memory.internal:23000", "POLIGN_API_KEY": key},
+)
+```
+
+The agent's records live in their own collection next to the memory
+collection (`<collection>_agents`).
+
+## Development
+
+```bash
+cd python/recall-langgraph
+python -m pip install -e . pytest pytest-asyncio
+pytest tests/unit_tests -q                # a fake Recall client, no network
+pytest tests/integration_tests -v         # real polign-server and polign CLI
+```
+
+The integration tests locate the binaries like the other packages' tests do
+(`POLIGN_SERVER`, `POLIGN_SOURCE`, `POLIGN_SERVER_VERSION`, or the latest
+release download) and skip when none is found or when the CLI has no
+`polign mcp -agent`. Both suites drive real compiled graphs with a scripted
+chat model, so no model API key is needed.

recall_langgraph-0.1.0.dist-info/RECORD

@@ -0,0 +1,8 @@
+recall_langgraph/__init__.py,sha256=gabHg0Xv5jJXF-Kl13XBrr4gnGpLCMjMXlWeUCIYQ8M,757
+recall_langgraph/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+recall_langgraph/resume.py,sha256=j3dEckPH_OKf_0xALz3F9wbiOEzla-FLYVGz7Va4Aao,14189
+recall_langgraph/tools.py,sha256=BwoO4qdTb9r8OHZlhBKMaYYzorHsSCHw9s4Qmr6jID4,5426
+recall_langgraph-0.1.0.dist-info/METADATA,sha256=o0s94O8MLvsFdnILolKH0DEOWjgIDFBaogV9KmXGMnk,9596
+recall_langgraph-0.1.0.dist-info/WHEEL,sha256=W3fkpkm7-wf9vBI5Z-7s0eWkeM-spu78I8Neb98DeEg,87
+recall_langgraph-0.1.0.dist-info/licenses/LICENSE,sha256=ra9FdQ2xyykGuwkTiZDGe8MaUdF61EbNfS6xW16vQUY,11343
+recall_langgraph-0.1.0.dist-info/RECORD,,

recall_langgraph-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.32.4
+Root-Is-Purelib: true
+Tag: py3-none-any

recall_langgraph-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Polign, Inc.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.