fix(explore): keep multi-term backend files from being buried by a denser frontend layer#917
Open
colbymchenry wants to merge 1 commit into
Open
fix(explore): keep multi-term backend files from being buried by a denser frontend layer#917colbymchenry wants to merge 1 commit into
colbymchenry wants to merge 1 commit into
Conversation
…nser frontend layer codegraph_explore's file sort is primarily driven by Random-Walk-with-Restart graph-centrality mass, seeded from the query's text matches. In a cross-layer monorepo (an API server alongside a much larger, internally dense frontend that mirrors the same domain words), that mass skews to the bigger layer — so a backend service/handler that genuinely matches several query terms, even when it's the #1 search hit, sorts below hits=0 frontend files and gets truncated out of the response, and the agent reads it back. Add a corroboration tier above the graph signal: a file that is BOTH an entry/central file AND matched by >=2 distinct query terms is kept in. The entry/central guard prevents an incidental multi-term file (a type/util file that isn't the flow) from displacing a graph-central answer file — a blunt hits-only tier regressed that case. Single-layer repos are unaffected. Gated by CODEGRAPH_RANK_NO_MULTITERM=1. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
codegraph_explorenow keeps a query-relevant backend file in the response when a larger, denser frontend layer would otherwise crowd it out of a cross-layer monorepo.Why
The explore file sort is primarily driven by Random-Walk-with-Restart (RWR) graph-centrality mass, seeded from the query's text matches. In a cross-layer monorepo — e.g. an
api/server alongside a much larger, internally denseapp/frontend that mirrors the same domain words — RWR mass concentrates in the bigger layer. A backend service/handler that matches several distinct query terms (often the #1 search hit, with many callers) is call-isolated from the frontend seed cluster, so it accrues little RWR mass, sorts belowhits=0frontend files, and gets truncated out of the response — and the agent reads the file back.How
A multi-term corroboration tier above the graph signal: a file that is BOTH (a) an entry/central file (search root, named seed, or graph-central hub) AND (b) matched by ≥2 distinct query terms ranks above the pure graph ordering.
The entry/central guard is what makes it safe: an incidental multi-term file that is neither entry nor central (a type/util file matching a common word but not on the flow) is NOT promoted, so it can't displace a graph-central answer file — a blunt
hits≥2-only tier regressed exactly that case (it floated unrelated type/util files above the real central renderer in a React render flow). Single-layer repos with one cluster are unaffected. Gated byCODEGRAPH_RANK_NO_MULTITERM=1.Validation
Deterministic, on real indexes (directus / n8n / excalidraw):
api/service moves from absent/mentioned to sourced in the explore output (was buried underapp/*.vue+ unrelated packages).__tests__/explore-corroboration-ranking.test.ts.🤖 Generated with Claude Code