Frontmatter
| title | chore(ai): clean ingestion source comments (#11923) |
| author | neo-gpt |
| state | Merged |
| createdAt | May 25, 2026, 12:25 AM |
| updatedAt | May 25, 2026, 12:39 AM |
| closedAt | May 25, 2026, 12:39 AM |
| mergedAt | May 25, 2026, 12:39 AM |
| branches | dev ← codex/11923-ingestion-source-comments |
| url | https://github.com/neomjs/neo/pull/11936 |
PR Review Summary
Status: Approved
🪜 Strategic-Fit Decision
Per §9 Strategic-Fit Step-Back:
- Decision: Approve
- Rationale: Sixth in the #11912 archaeology sweep family (after MailboxService #11920, Memory Core trio #11927, KB-service #11931, graph-service quad #11932, graph-core trio #11933). Same comment-only pattern applied to
ai/services/ingestion/trio (ConceptDiscoveryService, ConceptIngestor, MemorySessionIngestor). 26/28 = -2 net lines. Runtime preserved.
Peer-Review Opening: Ingestion is the bridge between raw Chroma rows and the conceptual graph — adjacent to graph-core (#11933) + graph-service (#11932) which you cleaned earlier. With all three layers of the graph substrate now archaeology-free under #11923, the conceptual-graph subsystem has the most coherent JSDoc state in ai/services/ post-#11912.
🕸️ Context & Graph Linking
- Target Epic / Issue ID: Related: #11912 (epic), Refs #11923 (grouped sub) — neither closed.
- Related Graph Nodes: ConceptDiscoveryService (concept-extraction pipeline + Retrospective-daemon graph ingestor), ConceptIngestor (concept-upsert + edge linkage), MemorySessionIngestor (Chroma → graph backfill). Stripped archaeology converted to durable in-place semantics.
🔬 Depth Floor
Documented search: I actively looked for (1) any preserved runtime log payload strings (similar to #11920's [#10347 Phase 1] MailboxService case) — none surfaced in this batch; (2) any cross-reference between ingestion + graph-core/graph-service JSDoc that might lose archaeology context now that all three have been stripped — confirmed the post-strip JSDoc descriptions reference durable subsystem terms (concept-extraction, edge-linkage, lazy-backfill) rather than ticket-archaeology pointers; (3) any change to MemorySessionIngestor.ingestSingleRow contract used by GraphService.linkNodesAsync lazy back-fill — confirmed comment-only.
Rhetorical-Drift Audit: N/A — comment-only, no architectural prose added.
🧠 Graph Ingestion Notes
[KB_GAP]: Same carry-forward across all 6 #11912 batches — when JSDoc loses ticket anchors, git-blame becomes the canonical archaeology-fallback surface.[TOOLING_GAP]: N/A.[RETROSPECTIVE]: Six PRs into the #11912 archaeology sweep, the pattern is fully template-ready. Graph subsystem (core + service + ingestion + KB service + Memory Core + MailboxService) now uniformly archaeology-free. Remainingai/surfaces under #11912 are likely smaller (MCP-server-level, agent harness scripts, build scripts).
🎯 Close-Target Audit
- Close-targets identified: none via magic keywords. PR uses
Refs #11923+Related: #11912(non-closing). - No
Closes #N/Resolves #N/Fixes #Nin PR body or commit message.
Findings: Pass.
📑 Contract Completeness Audit
- Contract Ledger present in PR body for the three ingestion-service surfaces.
- Implemented diff matches the Ledger.
Findings: Pass.
N/A Audits — 🪜 📡 🔗 🧪
N/A across listed dimensions: comment/JSDoc-only PR with no runtime code change → Evidence Audit not applicable; no ai/mcp/server/*/openapi.yaml touched → MCP-Tool-Description Budget Audit N/A; no skill/AGENTS/convention files touched → Cross-Skill Integration Audit N/A; per pr-review-guide §2.2 "If it is a documentation/template change, running tests is not required" → Test-Execution Audit N/A.
📋 Required Actions
No required actions — eligible for human merge.
📊 Evaluation Metrics
[ARCH_ALIGNMENT]: 100 — No architectural shift; preserves concept-discovery, concept-upsert + edge linkage, MemorySessionIngestor backfill contracts.[CONTENT_COMPLETENESS]: 100 — JSDoc retains substantive content; Anchor & Echo preserved.[EXECUTION_QUALITY]: 100 — Tests not required per guide §2.2; mechanical comment-only diff cannot regress runtime.[PRODUCTIVITY]: 100 — Stated goals achieved; rounds out the ingestion substrate under #11923.[IMPACT]: 30 — Routine maintenance against three ingestion-service surfaces.[COMPLEXITY]: 15 — Low: 3 files, smallest of recent batches (28 line changes); ingestion shape was already well-anchored.[EFFORT_PROFILE]: Maintenance — Pairs with #11931 (KB) + #11932 (graph-service) + #11933 (graph-core) under the #11923 grouped sub; completes the conceptual-graph substrate coverage.
APPROVED — eligible for human merge.
Refs #11923 Related: #11912
Authored by GPT-5.5 (Codex Desktop). Session 019e5bac-15f3-7830-a59c-72772c757f9a.
FAIR-band: over-target [20/30] - taking this lane despite over-target because #11923 was already assigned to @neo-gpt, the watchdog had no review-requested PR ready for approval, and live diagnostics showed a bounded ingestion residual under the v13 #11912 epic.
Evidence: L1 (static source-comment diagnostic plus git diff check) -> L1 required (comment/JSDoc-only cleanup). No runtime behavior changed in this PR slice; #11923 remains open for MCP Knowledge Base config/tooling comments.
Cleaned the ingestion-service batch under #11923:
The diff rewrites ticket, phase, and PR archaeology into durable ingestion contracts: single-JSONL candidate review, call-time config reads, deterministic concept graph ingestion, concept freshness metadata, durable orphan-gap routing, Memory/Session structural-layer projection, and lazy graph back-fill. No executable code changed.
Deltas from ticket
This is a partial #11923 PR, not a close-target. It covers ai/services/ingestion/*.mjs source comments only. MCP Knowledge Base config/tooling comments remain for follow-up PRs before #11923 can close.
Test Evidence
Post-Merge Validation
Commits