From 411f586c6710d8fc4f7f7b4cb55aab0f4b68bc70 Mon Sep 17 00:00:00 2001
From: Teknium <127238744+teknium1@users.noreply.github.com>
Date: Thu, 30 Apr 2026 03:32:14 -0700
Subject: [PATCH] refactor(gateway): extract _float_env helper for env-var
 float casts

Follow-up to the try/except guards added in the previous commit.
Four sibling call sites all read HERMES_AGENT_TIMEOUT /
HERMES_AGENT_TIMEOUT_WARNING / HERMES_AGENT_NOTIFY_INTERVAL via the
same read-env-or-fallback pattern, so factor it into _float_env(name,
default) alongside the existing _auto_continue_freshness_window()
helper.
---
 gateway/run.py | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

diff --git a/gateway/run.py b/gateway/run.py
index 8c6c4a51..af654021 100644
--- a/gateway/run.py
+++ b/gateway/run.py
@@ -122,6 +122,21 @@ def _auto_continue_freshness_window() -> float:
         return float(_AUTO_CONTINUE_FRESHNESS_SECS_DEFAULT)
 
 
+def _float_env(name: str, default: float) -> float:
+    """Read an env var as float, falling back to ``default`` on typos/empty.
+
+    A misconfigured env var (e.g. ``HERMES_AGENT_TIMEOUT=abc``) must not
+    crash the gateway or an agent turn.  Unset/empty also falls back.
+    """
+    raw = os.environ.get(name)
+    if raw is None or raw == "":
+        return float(default)
+    try:
+        return float(raw)
+    except (TypeError, ValueError):
+        return float(default)
+
+
 def _is_fresh_gateway_interruption(
     value: Any,
     *,
@@ -3948,10 +3963,7 @@ class GatewayRunner:
         # wall-clock age alone isn't sufficient.  Evict only when the agent
         # has been *idle* beyond the inactivity threshold (or when the agent
         # object has no activity tracker and wall-clock age is extreme).
-        try:
-            _raw_stale_timeout = float(os.getenv("HERMES_AGENT_TIMEOUT", 1800))
-        except (ValueError, TypeError):
-            _raw_stale_timeout = 1800.0
+        _raw_stale_timeout = _float_env("HERMES_AGENT_TIMEOUT", 1800)
         _stale_ts = self._running_agents_ts.get(_quick_key, 0)
         if _quick_key in self._running_agents and _stale_ts:
             _stale_age = time.time() - _stale_ts
@@ -11758,10 +11770,7 @@ class GatewayRunner:
         # Config: agent.gateway_notify_interval in config.yaml, or
         # HERMES_AGENT_NOTIFY_INTERVAL env var.  Default 180s (3 min).
         # 0 = disable notifications.
-        try:
-            _NOTIFY_INTERVAL_RAW = float(os.getenv("HERMES_AGENT_NOTIFY_INTERVAL", 180))
-        except (ValueError, TypeError):
-            _NOTIFY_INTERVAL_RAW = 180.0
+        _NOTIFY_INTERVAL_RAW = _float_env("HERMES_AGENT_NOTIFY_INTERVAL", 180)
         _NOTIFY_INTERVAL = _NOTIFY_INTERVAL_RAW if _NOTIFY_INTERVAL_RAW > 0 else None
         _notify_start = time.time()
 
@@ -11809,15 +11818,9 @@ class GatewayRunner:
             # Config: agent.gateway_timeout in config.yaml, or
             # HERMES_AGENT_TIMEOUT env var (env var takes precedence).
             # Default 1800s (30 min inactivity).  0 = unlimited.
-            try:
-                _agent_timeout_raw = float(os.getenv("HERMES_AGENT_TIMEOUT", 1800))
-            except (ValueError, TypeError):
-                _agent_timeout_raw = 1800.0
+            _agent_timeout_raw = _float_env("HERMES_AGENT_TIMEOUT", 1800)
             _agent_timeout = _agent_timeout_raw if _agent_timeout_raw > 0 else None
-            try:
-                _agent_warning_raw = float(os.getenv("HERMES_AGENT_TIMEOUT_WARNING", 900))
-            except (ValueError, TypeError):
-                _agent_warning_raw = 900.0
+            _agent_warning_raw = _float_env("HERMES_AGENT_TIMEOUT_WARNING", 900)
             _agent_warning = _agent_warning_raw if _agent_warning_raw > 0 else None
             _warning_fired = False
             _executor_task = asyncio.ensure_future(