hermes-agent/gateway/platforms
r266-tech 46f45104c4 fix(gateway): don't mark an entire chat dead on thread/message-level not_found
#55115 added the dead-target registry so confirmed-dead delivery targets are
short-circuited. Its documented scope (gateway/dead_targets.py) is deliberately
narrow: only *whole-chat* deaths -- the `forbidden` and chat-level `not_found`
(`chat not found`) kinds -- should be recorded; "Thread/topic-level not_found is
NOT recorded here ... a deleted topic does not mean the parent chat is dead."

But the implementation doesn't honor that scope. classify_send_error collapses
chat-level "chat not found" AND thread/message-level not_found ("thread not
found", "topic_deleted", "message_id_invalid", "message to edit/reply not
found") into one "not_found" kind, _DEAD_ERROR_KINDS contains "not_found"
wholesale, and deliver()'s except marks the PARENT chat_id dead. So a single
deleted Telegram topic or edited-away message permanently marks the entire chat
(and every future scheduled / cron / agent delivery to it) dead -- silently. The
adapter self-heal the docstring relies on only covers the non-private-group
thread retry; named-DM-topic and message-level failures propagate to deliver()'s
except and wrongly kill the whole chat.

Add is_chat_level_not_found() (factoring the not_found substrings into chat-level
vs sub-chat-level constants) and gate the delivery dead-path: a "not_found" only
marks the target dead when it is chat-level. classify_send_error's public
contract is unchanged (still returns "not_found" for every shape); only the
mark_dead decision is refined, restoring the registry's documented scope.

Cross-platform: telegram/slack/discord delivery all flow through
classify_send_error -> mark_dead. Adds regression tests through the real
deliver() path plus helper/classifier units.
2026-07-01 15:01:33 +05:30
..
qqbot fix(qqbot): stop 100% CPU spin when WebSocket is closed but not None (#31193, #31771) (#40574) 2026-06-06 18:44:44 -07:00
__init__.py perf(gateway): defer QQAdapter and YuanbaoAdapter imports via PEP 562 (#22790) 2026-05-09 13:17:48 -07:00
_http_client_limits.py fix(gateway): tighten httpx keepalive and close whatsapp typing-response leak (#18451) 2026-05-02 02:23:37 -07:00
ADDING_A_PLATFORM.md feat(whatsapp): add WhatsApp Business Cloud API adapter 2026-05-23 01:07:01 -04:00
api_server.py fix(api_server): pop fallback model kwarg to prevent AIAgent collision 2026-07-01 01:26:27 -07:00
base.py fix(gateway): don't mark an entire chat dead on thread/message-level not_found 2026-07-01 15:01:33 +05:30
bluebubbles.py fix(bluebubbles): drop unused with=participants from chat query 2026-07-01 00:42:56 -07:00
helpers.py refactor: extract shared GFM table→bullet helpers into helpers.py 2026-06-27 03:57:24 +05:30
msgraph_webhook.py fix(telegram): preserve Bot API update queue on watcher reconnect 2026-06-25 21:29:57 -07:00
signal.py fix(telegram): preserve Bot API update queue on watcher reconnect 2026-06-25 21:29:57 -07:00
signal_format.py fix(signal): share markdown formatting across send paths 2026-06-20 13:47:14 +05:30
signal_rate_limit.py feat(gateway/signal): add support for multiple images sending 2026-04-30 04:28:08 -07:00
webhook.py fix(gateway): sanitize agent error messages, validate webhook gh args 2026-06-28 18:53:26 -07:00
weixin.py fix(telegram): preserve Bot API update queue on watcher reconnect 2026-06-25 21:29:57 -07:00
whatsapp_cloud.py fix(whatsapp_cloud): resolve reply-to text so the agent sees reply context (#52957) 2026-06-26 01:05:05 -07:00
whatsapp_common.py fix(whatsapp): resolve phone↔LID aliases in adapter DM/group allowlist (#53588) 2026-06-27 04:17:12 -07:00
yuanbao.py fix(telegram): preserve Bot API update queue on watcher reconnect 2026-06-25 21:29:57 -07:00
yuanbao_media.py fix(security): SSRF guard yuanbao media download_url (#54470) 2026-06-28 15:29:59 -07:00
yuanbao_proto.py feat(Yuanbao): support wechat forward msg (#43508) 2026-06-12 02:06:47 -07:00
yuanbao_sticker.py yuanbao platform (#16298) 2026-04-26 18:50:49 -07:00