fix: isolate memory flush messages from main agent history - #7
Merged
Conversation
The flush prompt and its tool call messages were pushed into the main history and newMessages arrays. After compaction kept recent messages, the flush prompt became the last user message — causing the LLM to respond to it instead of the original request, leaking flush text to the user via Telegram. The flush now operates on a separate flushHistory copy. Flush messages are persisted to the session store for audit but never touch the working history or newMessages. Compaction sees only the clean original conversation. https://claude.ai/code/session_01Ac1E6X4Yw9Bt3iyktMoS5C
…kens Since compactAndSync now receives the clean original history (without flush messages), the token count should reflect the original history size. The flushTokens variable is no longer needed. https://claude.ai/code/session_01Ac1E6X4Yw9Bt3iyktMoS5C
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The flush prompt and its tool call messages were pushed into the main
history and newMessages arrays. After compaction kept recent messages,
the flush prompt became the last user message — causing the LLM to
respond to it instead of the original request, leaking flush text to
the user via Telegram.
The flush now operates on a separate flushHistory copy. Flush messages
are persisted to the session store for audit but never touch the
working history or newMessages. Compaction sees only the clean
original conversation.
https://claude.ai/code/session_01Ac1E6X4Yw9Bt3iyktMoS5C