fix(mem0-agent): non-blocking stdin, clean session_state, lint

Found by running the CLI live against a scratch project:
- hook_input() blocked forever on an open stdin pipe with no data. A real hook
  writes JSON and closes the pipe, but any manual invocation hung; now polled
  with select() and never blocks.
- the open-thread snapshot was filling with progress lines and file-modification
  lists -- v1's exact failure -- so it now filters turns through the same
  hard-drop rules the capture gate uses.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

integrations/mem0-agent/src/mem0_agent/cli.py

@@ -24,12 +24,32 @@ from .settings import CAPTURE_LEVELS, MEMORY_MODES, RETRIEVAL_LEVELS, Settings
 # --------------------------------------------------------------------------
 # hook plumbing
 # --------------------------------------------------------------------------
-def hook_input() -> dict:
+def hook_input(timeout: float = 0.25) -> dict:
-    """Editors hand hooks a JSON payload on stdin. Absent or malformed is fine."""
+    """Editors hand hooks a JSON payload on stdin. Absent or malformed is fine.
-    if sys.stdin is None or sys.stdin.isatty():
+
+    Never blocks. A hook's stdin is written and closed, but a manual invocation inherits
+    an open pipe with nothing in it -- reading that would hang the command forever.
+    """
+    stream = sys.stdin
+    if stream is None or not hasattr(stream, "read"):
         return {}
     try:
-        raw = sys.stdin.read()
+        if stream.isatty():
+            return {}
+    except Exception:
+        return {}
+
+    # StringIO and friends (tests) are readable immediately; real pipes get a poll.
+    if hasattr(stream, "fileno"):
+        try:
+            import select
+
+            if not select.select([stream], [], [], timeout)[0]:
+                return {}
+        except Exception:
+            pass  # not selectable (e.g. StringIO) -- fall through and read
+    try:
+        raw = stream.read()
         return json.loads(raw) if raw.strip() else {}
     except Exception:
         return {}

integrations/mem0-agent/src/mem0_agent/maintain.py

@@ -12,7 +12,7 @@ from __future__ import annotations
 import re
 from dataclasses import dataclass, field
 from itertools import combinations
-from typing import Any, Iterable
+from typing import Iterable
 
 from .api import expiry_date, results_of
 from .config import filters as F

integrations/mem0-agent/src/mem0_agent/settings.py

@@ -231,7 +231,7 @@ class SessionState:
 
     def read_lines(self, name: str) -> list[dict]:
         try:
-            return [json.loads(l) for l in self._p(name).read_text().splitlines() if l.strip()]
+            return [json.loads(line) for line in self._p(name).read_text().splitlines() if line.strip()]
         except Exception:
             return []
 

integrations/mem0-agent/src/mem0_agent/transcript.py

@@ -86,10 +86,35 @@ def windows_since(turns: list[dict], processed: int, size: int = 4) -> list[list
     return [fresh[i:i + size] for i in range(0, len(fresh), size) if fresh[i:i + size]]
 
 
+MIN_THREAD_WORDS = 4
+
+
+def _signal_turns(turns: list[dict]) -> list[dict]:
+    """Drop the same mechanical noise the capture gate drops.
+
+    Without this the open-thread snapshot fills up with progress lines and
+    "I modified VERSION, cli.py, ..." -- the exact content the audit found made v1's
+    session summaries worthless.
+    """
+    from .triggers import _turn_drop_reason, natural_words
+
+    kept = []
+    for t in turns:
+        if t.get("tool_only") or not t.get("content"):
+            continue
+        if _turn_drop_reason([t]):
+            continue
+        if natural_words(t["content"]) < MIN_THREAD_WORDS:
+            continue  # "ok", "good catch" -- acknowledgements, not state
+        kept.append(t)
+    return kept
+
+
 def summarize_open_thread(turns: list[dict], limit: int = 3) -> str:
     """A plain-language snapshot for session_state: what we were doing, where it stopped."""
-    users = [t["content"] for t in turns if t["role"] == "user" and t["content"]]
+    signal = _signal_turns(turns)
-    assistants = [t["content"] for t in turns if t["role"] == "assistant" and t["content"]]
+    users = [t["content"] for t in signal if t["role"] == "user"]
+    assistants = [t["content"] for t in signal if t["role"] == "assistant"]
     if not users and not assistants:
         return ""
     goal = users[0][:300] if users else ""

integrations/mem0-agent/tests/test_integration.py

@@ -313,3 +313,26 @@ def test_full_session_produces_few_memories(tmp_path):
     assert summary["sent"] >= 1
     for added in ctx.api.added:
         assert "epoch" not in added["messages"][0]["content"].lower()
+
+
+# --------------------------------------------------------------------------
+# session_state quality
+# --------------------------------------------------------------------------
+def test_open_thread_excludes_mechanical_noise():
+    """v1's session summaries were file lists. The snapshot must carry intent instead."""
+    turns = [
+        {"role": "assistant", "content": HEARTBEAT, "tool_only": False},
+        {"role": "user", "content": "Let's get the sandbox e2e suite passing today.", "tool_only": False},
+        {"role": "user", "content": "ok", "tool_only": False},
+        {"role": "assistant", "content": FILE_LIST, "tool_only": False},
+    ]
+    thread = transcript.summarize_open_thread(turns)
+    assert "sandbox e2e" in thread
+    assert "VERSION" not in thread and "chat.py" not in thread
+    assert "epoch" not in thread.lower()
+
+
+def test_open_thread_is_empty_when_there_is_only_noise():
+    turns = [{"role": "assistant", "content": HEARTBEAT, "tool_only": False},
+             {"role": "assistant", "content": FILE_LIST, "tool_only": False}]
+    assert transcript.summarize_open_thread(turns) == ""

integrations/mem0-agent/tests/test_maintain.py

@@ -1,6 +1,5 @@
 """Consolidation must never lose data, and must never touch a pinned memory."""
 
-import pytest
 
 from mem0_agent import maintain
 from mem0_agent.maintain import jaccard, tokens