Clamp packet ranges and make handoff metadata optional - #27
Merged
Merged
Conversation
…tory_packet Retrieval measured from 98 recorded cells without model calls shows every arm usually retrieves the required evidence; the Context arm's friction is tool errors (one call in five on DeepSeek). Most were whole-file reads with a guessed endLine and placeholder packet metadata.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Model-free retrieval analysis of the 98 recorded experiment cells (
docs/experiments/retrieval-recall-20260923/):repository_packetthrough MCP (CLI spec unchanged):endLinepast the end of a file reads to its last line;originalSpecrecords the range read soverifyPacketrebuilds it exactly; a range starting past the end is still rejected;sourcesis required; blank or omitted handoff metadata gets a neutral task and acceptance check;mode: "build".Checks:
npm run check(182 context-tools tests), navigation smoke, both benchmark gates with recall 1.0.🤖 Generated with Claude Code
https://claude.ai/code/session_01KzXbuBSq88i3Re82qt6c4L