common/hermes
2
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix(slack): extract rich_text quotes/lists and link unfurl previews

Browse Source 
Slack's modern composer sends messages with a 'blocks' array that
contains rich_text elements. When a user forwards or quotes another
message, the quoted content shows up in the rich_text_quote children
of that array — and is NOT included in the plain 'text' field. The
agent saw only the lossy plain text and was blind to forwarded /
quoted content. Same story for link unfurl previews (Notion, docs,
GitHub, etc.) which Slack puts in the 'attachments' array.

Two fixes in the inbound handler:

1. _extract_text_from_slack_blocks walks rich_text / rich_text_quote /
   rich_text_list / rich_text_preformatted trees and renders readable
   text ('> quoted', '• bullet', code fences), dedupes against the
   plain text field, and appends the extracted content so the agent
   sees everything.

2. Link unfurl / attachment preview extraction reads title, url,
   body, and footer from the 'attachments' array and appends a
   '📎 [title](url)\n   body\n   _footer_' section per preview.
   Skips is_msg_unfurl to avoid echoing our own Slack replies back.

Routing is careful not to trust augmented text: mention gating
(is_mentioned) and slash-command detection both run against the
original 'text' field, so forwarded content containing '<@bot>' or
'/deploy' in a quote can't trick the bot into responding in a
channel it shouldn't or classifying a normal message as a command.

Adjustment from original PR: dropped _serialize_slack_blocks_for_agent,
which inlined a redacted JSON dump of non-rich_text blocks (section,
accessory, actions, etc.) — the agent would see the raw Block Kit
structure for UI-heavy alerts. It added up to 6000 characters to the
prompt context on every qualifying message with no opt-out. The
rich_text extraction and attachment unfurls cover the common bug-fix
case (quoted/forwarded content + link previews) without the prefill
tax. If a user needs block inspection later, it can return as a
config opt-in.

Also updates the Slack platform notes in session.py to accurately
describe what the gateway inlines.
This commit is contained in:
Wang-tianhao
2026-04-26 13:02:27 -07:00
committed by Teknium
parent 4921b26945
commit 6087e04043
4 changed files with 429 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
tests/gateway/test_session.py
View File

@@ -245,6 +245,7 @@ class TestBuildSessionContextPrompt:
assert "Slack" in prompt
assert "cannot search" in prompt.lower()
assert "pin" in prompt.lower()
assert "current message's slack block/attachment payload" in prompt.lower()
def test_discord_prompt_with_channel_topic(self):
"""Channel topic should appear in the session context prompt."""

178
tests/gateway/test_slack.py
View File

@@ -355,15 +355,17 @@ class TestSendVideo:
# ---------------------------------------------------------------------------
class TestIncomingDocumentHandling:
def _make_event(self, files=None, text="hello", channel_type="im"):
def _make_event(self, files=None, text="hello", channel_type="im", blocks=None, attachments=None):
"""Build a mock Slack message event with file attachments."""
return {
"text": text,
"user": "U_USER",
"channel": "C123",
"channel": "D123",
"channel_type": channel_type,
"ts": "1234567890.000001",
"files": files or [],
"blocks": blocks or [],
"attachments": attachments or [],
}
@pytest.mark.asyncio
@@ -540,6 +542,178 @@ class TestIncomingDocumentHandling:
assert "403" in msg_event.text
assert "what's in this?" in msg_event.text
@pytest.mark.asyncio
async def test_rich_text_blocks_do_not_duplicate_plain_text(self, adapter):
"""Plain rich_text composer blocks match the plain text field exactly,
so the dedupe guard keeps the message clean."""
event = self._make_event(
text="hello world",
blocks=[
{
"type": "rich_text",
"elements": [
{
"type": "rich_text_section",
"elements": [
{"type": "text", "text": "hello world"},
],
}
],
}
],
)
await adapter._handle_slack_message(event)
msg_event = adapter.handle_message.call_args[0][0]
assert msg_event.text == "hello world"
@pytest.mark.asyncio
async def test_rich_text_quotes_and_lists_are_extracted(self, adapter):
"""Nested quote and list content should be surfaced from rich_text blocks."""
event = self._make_event(
text="Can you summarize this?",
blocks=[
{
"type": "rich_text",
"elements": [
{
"type": "rich_text_quote",
"elements": [
{
"type": "rich_text_section",
"elements": [{"type": "text", "text": "Quoted line"}],
}
],
},
{
"type": "rich_text_list",
"style": "bullet",
"elements": [
{
"type": "rich_text_section",
"elements": [{"type": "text", "text": "First bullet"}],
},
{
"type": "rich_text_section",
"elements": [{"type": "text", "text": "Second bullet"}],
},
],
},
],
}
],
)
await adapter._handle_slack_message(event)
msg_event = adapter.handle_message.call_args[0][0]
assert "Can you summarize this?" in msg_event.text
assert "> Quoted line" in msg_event.text
assert "• First bullet" in msg_event.text
assert "• Second bullet" in msg_event.text
@pytest.mark.asyncio
async def test_attachments_unfurl_text_is_appended_even_when_url_is_in_message(self, adapter):
"""Shared URLs should still expose unfurl preview text to the agent."""
event = self._make_event(
text="Look at this doc https://example.com/spec",
attachments=[
{
"title": "Spec",
"from_url": "https://example.com/spec",
"text": "The latest product spec preview",
"footer": "Notion",
}
],
)
await adapter._handle_slack_message(event)
msg_event = adapter.handle_message.call_args[0][0]
assert "Look at this doc https://example.com/spec" in msg_event.text
assert "📎 [Spec](https://example.com/spec)" in msg_event.text
assert "The latest product spec preview" in msg_event.text
assert "_Notion_" in msg_event.text
@pytest.mark.asyncio
async def test_message_unfurl_attachments_are_skipped(self, adapter):
"""Message unfurls should be skipped to avoid echoing Slack message copies."""
event = self._make_event(
text="https://example.com/thread",
attachments=[
{
"is_msg_unfurl": True,
"title": "Thread copy",
"text": "This should not be appended",
}
],
)
await adapter._handle_slack_message(event)
msg_event = adapter.handle_message.call_args[0][0]
assert msg_event.text == "https://example.com/thread"
@pytest.mark.asyncio
async def test_channel_routing_ignores_bot_mentions_inside_block_text(self, adapter):
"""Block-extracted text with a bot mention must not satisfy mention
gating in channels — routing decisions use the original user text so
quoted/forwarded content can't trick the bot into responding."""
event = self._make_event(
text="please review",
channel_type="channel",
blocks=[
{
"type": "rich_text",
"elements": [
{
"type": "rich_text_quote",
"elements": [
{
"type": "rich_text_section",
"elements": [{"type": "text", "text": "Contains <@U_BOT> in quoted text"}],
}
],
}
],
}
],
)
await adapter._handle_slack_message(event)
adapter.handle_message.assert_not_called()
@pytest.mark.asyncio
async def test_quoted_slash_command_text_does_not_change_message_type(self, adapter):
"""Quoted slash-like content should not convert a normal message into a command."""
event = self._make_event(
text="",
blocks=[
{
"type": "rich_text",
"elements": [
{
"type": "rich_text_quote",
"elements": [
{
"type": "rich_text_section",
"elements": [{"type": "text", "text": "/deploy now"}],
}
],
}
],
}
],
)
await adapter._handle_slack_message(event)
msg_event = adapter.handle_message.call_args[0][0]
assert msg_event.message_type == MessageType.TEXT
assert "> /deploy now" in msg_event.text
# ---------------------------------------------------------------------------
# TestMessageRouting