Make glob/grep truncation explicit and use the calibrated estimate in /compact - #8
Open
uuzzrm wants to merge 1 commit into
Open
Make glob/grep truncation explicit and use the calibrated estimate in /compact#8uuzzrm wants to merge 1 commit into
uuzzrm wants to merge 1 commit into
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two small honesty fixes, both about numbers the agent sees.
globandgrepboth stop walking at a file cap (20k / 5k), but neither said anything about it. On a big repo the model gets a result that just... ends, and reasonably concludes there are no more files. They now append an explicit "search capped at N files, results truncated" note — and grep says the search was capped even when nothing matched, so a truncated search can't masquerade as "no matches"./compactdisplayed the raw char-based estimate while the compressor itself decides on the calibrated measure — two numbers that could disagree by a lot after a real conversation. It now usescontext.measure()for before/after.Truncation needs >20k files to trigger, so I couldn't write a cheap regression test for that branch — added normal-path tests for both tools instead so the refactor itself is covered.