feat(telegram): handle incoming user reactions as feedback signals - #23731
Open
zetxek wants to merge 1 commit into
Open
feat(telegram): handle incoming user reactions as feedback signals#23731zetxek wants to merge 1 commit into
zetxek wants to merge 1 commit into
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR adds support in the Telegram gateway adapter for interpreting incoming emoji reactions on bot messages as user feedback signals, persisting them to disk for later analysis/learning.
Changes:
- Registers a Telegram
MessageReactionHandler(when available) to receive reaction updates. - Implements
TelegramAdapter._handle_reaction()to map supported emojis to feedback types and append entries to a JSONL log plus an additional Markdown log for stronger signals. - Adds a new pytest suite covering reaction handling and file output behavior.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 8 comments.
| File | Description |
|---|---|
gateway/platforms/telegram.py |
Adds handler registration and a new _handle_reaction() method to log reaction feedback to disk. |
tests/gateway/test_telegram_incoming_reactions.py |
Introduces tests for reaction-to-feedback mapping, ignored cases, multi-reaction updates, and file writes. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Comment on lines
+4614
to
+4621
| feedback_file = pathlib.Path.home() / ".hermes" / "feedback.jsonl" | ||
| feedback_file.parent.mkdir(parents=True, exist_ok=True) | ||
| with open(feedback_file, "a") as f: | ||
| f.write(json.dumps(entry) + "\n") | ||
|
|
||
| # For negative feedback or saves, write to memory log | ||
| if feedback_type in ("negative", "save", "strong_positive"): | ||
| memory_dir = pathlib.Path.home() / ".hermes" / "memory" |
Comment on lines
+4614
to
+4618
| feedback_file = pathlib.Path.home() / ".hermes" / "feedback.jsonl" | ||
| feedback_file.parent.mkdir(parents=True, exist_ok=True) | ||
| with open(feedback_file, "a") as f: | ||
| f.write(json.dumps(entry) + "\n") | ||
|
|
Comment on lines
+4616
to
+4618
| with open(feedback_file, "a") as f: | ||
| f.write(json.dumps(entry) + "\n") | ||
|
|
Comment on lines
+4620
to
+4624
| if feedback_type in ("negative", "save", "strong_positive"): | ||
| memory_dir = pathlib.Path.home() / ".hermes" / "memory" | ||
| memory_dir.mkdir(parents=True, exist_ok=True) | ||
| log_file = memory_dir / "feedback-log.md" | ||
| ts = datetime.datetime.utcnow().strftime("%Y-%m-%d %H:%M") |
Comment on lines
+4567
to
+4572
| Supported reactions and their meanings: | ||
| 👍 (\U0001f44d) → positive feedback | ||
| 👎 (\U0001f44e) → negative feedback — also logged to memory/feedback-log.md | ||
| ❤️ (\U00002764) → save/bookmark this response — also logged to memory | ||
| 🔥 (\U0001f525) → strong positive — also logged to memory | ||
|
|
Comment on lines
+4605
to
+4608
| entry = { | ||
| "timestamp": datetime.datetime.utcnow().isoformat(), | ||
| "emoji": emoji, | ||
| "feedback_type": feedback_type, |
| assert entry["emoji"] == "\U00002764" | ||
|
|
||
|
|
||
| @pytest.mark.asyncio |
Comment on lines
+140
to
+151
|
|
||
| @pytest.mark.asyncio | ||
| async def test_thumbs_down_writes_to_memory_log(tmp_path, monkeypatch): | ||
| """👎 reaction should write a note to memory/feedback-log.md.""" | ||
| monkeypatch.setattr(pathlib.Path, "home", lambda: tmp_path) | ||
| adapter = _make_adapter() | ||
| update = _make_reaction_update("\U0001f44e", message_id=77, chat_id=555) | ||
|
|
||
| await adapter._handle_reaction(update, MagicMock()) | ||
|
|
||
| log_file = tmp_path / ".hermes" / "memory" / "feedback-log.md" | ||
| assert log_file.exists() |
alt-glitch
added
type/feature
Contributor
1 task
Contributor
|
Thanks for the reaction-feedback contribution. The underlying gap remains: current Telegram startup has no inbound reaction handler ( Problems
Suggested changes
Automated hermes-sweeper review. |
teknium1
added
sweeper:risk-compatibility
|
Use case that triggered finding this issue: hydration reminders sent by a cron job, acknowledged by the user reacting with 👍. The reaction is the natural Telegram UX — no new habit to form. Would love to see this land! Note: I have implemented a local patch that stores incoming reactions to |
This was referenced Aug 4, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds support for receiving emoji reactions from users on Telegram as feedback signals, enabling self-learning over time.
How it works
When a user reacts to one of the bot's messages with a supported emoji, the bot:
MessageReactionUpdatedevent (already received viaUpdate.ALL_TYPES— no polling change needed)~/.hermes/feedback.jsonl~/.hermes/memory/feedback-log.mdfor the agent to learn fromEmoji → Feedback mapping
positivenegativesavestrong_positiveUnknown emojis are silently ignored.
Files changed
gateway/platforms/telegram.py— adds_handle_reaction()method and registersMessageReactionHandlerinside_start_pollingtests/gateway/test_telegram_incoming_reactions.py— 17 new tests covering all feedback types, edge cases (missing fields, unknown emojis, empty reactions), and file I/OTests
All 30 tests pass (17 new + 13 existing reaction tests):
Backwards compatibility
MessageReactionHandlerimport is wrapped intry/except ImportError— gracefully skips on older PTB versions with a debug log