feat: embed-only ingest + chunk-only ask (no-graph redesign, phase 1+2) - #87
Merged
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Phases 1 & 2 of the knowledge-service lazy-RAG redesign (design spec: aegis docs/superpowers/specs/2026-06-24-ks-lazy-rag-redesign-design.md).
Phase 1 — ingest is chunk + embed only.
ingestion/worker.pyruns EmbedPhase and stops. NLP pre-pass, LLM triple extraction, coreference, and process/graph phases are skipped (no triple_outbox writes, no inference, no provenance). This removes the per-item extraction load.Phase 2 —
askis chunk-only multi-doc synthesis.api/ask.pyalways retrieves top-k chunks and synthesizes with citations. Intent classification, graph routing, noisy-or confidence, triple evidence, and contradictions are gone.AskResponsekeepsanswer/sources; graph-only fields are empty stubs for backward compat.Not deleted yet (phase 3): ExtractPhase/ProcessPhase/coreference/inference/nlp/ontology and RAGRetriever graph methods remain in the tree, just unreachable from ingest/ask. All marked with
ponytail:comments.Tests: 33/33 pass (5 graph-behavior tests updated to chunk-only expectations).
ruff checkclean.Merging builds
:latestbut does NOT auto-deploy — deploy is a separateservice update --force aegis_knowledge.🤖 Generated with Claude Code