Repository navigation
Conversation
Co-authored-by: Codex <noreply@openai.com> Signed-off-by: Gil Birman <gil@squareup.com>
gilgtm
force-pushed
the
gil/agent-instruction-conflicts
branch
from
November 16, 2026 11:46
6223e98 to
2ce590f
Compare
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
The base prompt can require a new worktree even when a conversation already has a working folder. Its reporting rules can also conflict with plugin instructions.
What
Use assigned conversation folders and explicit project folders, follow their rules, and reuse the worktree for branch or PR follow-ups. Let authorized plugin tools route reports, show progress as status, and request a final message without duplicate callback posts.
Keep older saved workspace instructions readable, including Canvases near the size limit. This is the bottom PR; message-relaying attribution builds on it.
Risk
The shared base prompt affects existing harnesses. Workspace parsing still checks the exact managed text, and saving retains the current size limit.
Testing
Historical Canvas fixtures at exactly 256 KiB reproduce the readback failure before its repair. No browser journeys were added or removed. Native app and human acceptance remain deferred; this PR stays draft.
Not in this PR
Message-relaying attribution belongs to the top PR. Memory policy and Canvas trust precedence remain outside this change.
Generated with Codex