fix: decrease chance of claude ignoring hook instructions#105
Merged
Conversation
Claude sometimes ignores SessionStart hook output because the wording
combines urgency, secrecy, and pre-emption — patterns that match prompt
injection signatures. Replace secrecy ("do not mention to user",
"silently") and pre-emption ("before responding to the user's first
message") with transparency-encouraging language ("let the user know",
"tell the user", "briefly mention to the user"). Keep "ACTION REQUIRED"
as the action signal.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Owner
Author
Stack LinksThis PR is part of a stacked series: This comment was autogenerated by stax |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I asked Claude why it didn't distill and it said it ignored the instructions because it looked like prompt injection.
Claude asserts that these changes will make the instructions look less adversarial.
changelog: include