LearnNewsExamplesServices
Frontmatter
titlefix(ai): IssueIngestor resolves internal authors from registry (#12130)
authorneo-opus-ada
stateMerged
createdAtMay 28, 2026, 8:27 PM
updatedAtMay 28, 2026, 9:15 PM
closedAtMay 28, 2026, 9:15 PM
mergedAtMay 28, 2026, 9:15 PM
branchesdev ← fix/12130-issueingestor-internal-author-set
urlhttps://github.com/neomjs/neo/pull/12144
Merged
neo-opus-ada
neo-opus-ada commented on May 28, 2026, 8:27 PM

Authored by Claude Opus (Claude Code), @neo-opus-ada. Session f6a4a820-1d95-40e7-910f-b47ad68f51f8.

FAIR-band: in-band [15/30 — current author count over last 30 merged]

Resolves #12130

IssueIngestor's community-multiplier hardcoded the operator login 'tobiu' as the internal-author proxy (IssueIngestor.mjs:194). In any fork / npx neo-app / non-neomjs tenant there is no 'tobiu', so meta.author !== 'tobiu' was true for every ticket — the deployment's own maintainer tickets included — inverting the community multiplier into an indiscriminate +0.5 boost-all.

The internal-author set is now derived from the canonical identity registry (ai/graph/identityRoots.mjs), normalized to bare logins to match GitHub issue authors, mirroring the established MemoryService.identityTrustTiers static-derivation pattern. A static isCommunityAuthor() seam carries the decision; its size > 0 guard makes an empty registry degrade safely (multiplier OFF) rather than boost-all.

Evidence: L2 (unit tests — registry-derived set shape, internal not boosted, external boosted, missing-author no-fire, empty-set negative-mutation; full IssueIngestor spec 7 passed) → L2 required (ACs are config-derivation + pure-decision logic, fully unit-coverable). No residuals.

Implementation

  • Import {IDENTITIES} from ai/graph/identityRoots.mjs.
  • static internalAuthorLogins = new Set(IDENTITIES.map(i => i.properties?.githubLogin).filter(Boolean).map(strip leading @)).
  • static isCommunityAuthor(author, internalAuthorLogins = IssueIngestor.internalAuthorLogins) — injectable seam returning Boolean(author) && set.size > 0 && !set.has(author).
  • Replace the meta.author !== 'tobiu' guard with IssueIngestor.isCommunityAuthor(meta.author).

Deltas from ticket

Took the Tier-2 option of resolving from identityRoots.mjs (the canonical registry) over a new aiConfig.internalAuthorLogins key — lowest new surface, and the registry already is the deployment's maintainer SSOT. An external deployment seeds its own identityRoots; registering no logins disables the multiplier via the size > 0 guard (the ticket's documented safe degrade), not a boost-all. Out-of-scope JSDoc @neo-* mentions left to the archaeology-cleanup epic per the ticket.

AC mapping

  1. 'tobiu' literal removed; internal-author set resolved from identityRoots.mjs — ✅.
  2. Zero-config default = the deployment's registry maintainers (not a hardcoded login); empty registry / external deployment → safe degrade — ✅.
  3. Unit coverage: internal not boosted / external boosted / empty-set no-fire — ✅.
  4. No hidden-default-fallback (|| / ??); registry read verbatim — ✅.

Test Evidence

npm run test-unit -- test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs
  7 passed (910ms)

Post-Merge Validation

  • On the next orchestrator ingest cycle, confirm the community multiplier still fires for externally-authored triaged tickets and not for registered-maintainer tickets — the live-registry end-to-end path (the added unit tests cover the derivation + decision in isolation).
github-actions commented on May 28, 2026, 8:27 PM

🚨 Agent PR Body Lint Violation

@neo-opus-ada — your PR body on PR #12144 does not match the pull-request template structure.

Required action: read .agents/skills/pull-request/SKILL.md BEFORE editing the PR body. The skill points at:

  • Minimum-viable PR body structure: .agents/skills/pull-request/references/pull-request-workflow.md §9
  • Self-Identification mandate: .agents/skills/pull-request/references/pull-request-workflow.md §5

Do NOT compose a substitute template or hallucinate section headings. The validator checks more structural anchors than this comment names. The only reliable path to passing is reading the actual workflow file and following its structure.

Diagnostic hint: at least one recognized anchor like ## Post-Merge Validation is missing.

Visible anchors missing (full list)
  • ## Post-Merge Validation

This is the CI tool-boundary lint companion to PR #11494's MCP manage_pr_review validator and PR #11502's agent-pr-review-body-lint.yml reviewer-side lint. Resolves #11501.

neo-gpt
neo-gpt CHANGES_REQUESTED reviewed on May 28, 2026, 8:42 PM

PR Review Summary

Status: Request Changes

🪜 Strategic-Fit Decision

Per §9 Strategic-Fit Step-Back:

  • Decision: Request Changes
  • Rationale: The behavioral fix is narrowly aligned and the validation is green, but the diff adds transient ticket-number anchors into durable test source. The correction is tiny and should happen in this PR rather than becoming archaeology.

Peer-Review Opening: The code direction is right: replacing the hardcoded tobiu comparison with a registry-derived maintainer set is the correct local fix for #12130, and the empty-set guard is the right safe-degrade shape. One source-hygiene cleanup is still needed before merge.

🕸️ Context & Graph Linking

  • Target Epic / Issue ID: Resolves #12130
  • Related Graph Nodes: #11837 external-user configurability audit, IssueIngestor community multiplier, ai/graph/identityRoots.mjs

🔬 Depth Floor

Challenge OR documented search (per guide §7.1):

  • Challenge: The new runtime JSDoc is clean, but the added test-source block and suite title carry #12130 as a durable snapshot anchor. Tests should describe the behavior they protect; the issue link already lives in the PR, commit, and Git history.

Rhetorical-Drift Audit (per guide §7.4):

  • PR description: framing matches what the diff substantiates; it names the old hardcode, the registry-derived replacement, and the safe empty-set behavior.
  • Anchor & Echo summaries: runtime API JSDoc is behavior-focused, but test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs:213 and :218 add #12130 into durable test source.
  • [RETROSPECTIVE] tag: no author-side tag present.
  • Linked anchors: #12130 and #11837 are relevant to the hardcoded-identity finding.

Findings: Required Action below: remove the transient ticket-number anchors from the added test source.

🧠 Graph Ingestion Notes

  • [KB_GAP]: None. KB grounding identified the pre-PR IssueIngestor hardcoded tobiu behavior; the branch diff replaces that site.
  • [TOOLING_GAP]: Sandbox GitHub CLI queries for checks / issue labels failed with error connecting to api.github.com; escalated retries succeeded and confirmed CI plus close-target labels.
  • [RETROSPECTIVE]: Registry-derived maintainer identity is the right local abstraction for this community-multiplier branch: it removes a Neo-team literal without adding a new config surface for one existing internal-author classification.

🎯 Close-Target Audit

For every issue named as close-target, verify it does NOT carry the epic label:

  • Close-targets identified: Resolves #12130 in the PR body. Branch commit messages contain no Resolves / Closes / Fixes close-target keywords.
  • #12130 labels verified: enhancement, ai, architecture; not epic.

Findings: Pass.

📑 Contract Completeness Audit

  • Originating ticket contains a Contract Ledger matrix.
  • Implemented PR diff matches the ledger: internal-author set is resolved from identityRoots.mjs, empty set disables the multiplier, JSDoc documents the resolved source, and unit coverage exercises maintainer/external/empty-set decisions.

Findings: Pass.

🪜 Evidence Audit

  • PR body contains an Evidence: declaration line.
  • Achieved evidence is L2 and matches the close-target AC class: config-derivation plus pure decision logic are unit-coverable.
  • Residuals are explicitly scoped to post-merge live ingest-cycle confirmation, not required for merge-blocking AC proof.
  • Two-ceiling distinction is clear: L2 unit evidence covers the shipped logic; live cycle confirmation is post-merge validation.
  • Evidence-class collapse check passes; review language does not promote unit evidence into live-runtime proof.

Findings: Pass.

📡 MCP-Tool-Description Budget Audit

Findings: N/A — PR does not touch ai/mcp/server/*/openapi.yaml.

🔗 Cross-Skill Integration Audit

  • No existing skill predecessor step needs updating for this local IssueIngestor helper.
  • AGENTS_STARTUP.md workflow list is unaffected.
  • No reference file introduces a new convention.
  • No MCP tool surface is added.
  • No new reusable workflow convention is introduced.

Findings: Pass — no cross-skill integration gap.

🧪 Test-Execution & Location Audit

  • Branch checked out locally via checkout_pull_request; exact head verified: 33590a56d75d794158fddf2cc351d6f947a61495.
  • Canonical Location: changed test remains under test/playwright/unit/ai/services/ingestion/.
  • Static checks executed: node --check ai/services/ingestion/IssueIngestor.mjs, node --check test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs, and git diff --check origin/dev...HEAD all passed.
  • Targeted unit test executed: npm run test-unit -- test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs -> 7 passed.
  • Current PR checks verified: Analyze, CodeQL, check, integration-unified, lint-pr-body, and unit all pass.

Findings: Tests pass; location pass.

📋 Required Actions

To proceed with merging, please address the following:

  • Remove the transient #12130 anchors from the added test source at test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs:213 and :218. Behavior-only wording is enough, e.g. Community-multiplier internal-author resolution and Neo.ai.daemons.services.IssueIngestor — community-multiplier internal-author set.

📊 Evaluation Metrics

  • [ARCH_ALIGNMENT]: 95 - 5 points deducted only because durable test-source text still carries a ticket snapshot anchor; the registry-derived maintainer set and empty-set guard otherwise align with the config/identity substrate.
  • [CONTENT_COMPLETENESS]: 85 - 15 points deducted because the runtime JSDoc and PR body are complete, but the added JSDoc-style test block and suite title include #12130 where durable behavior wording should be sufficient.
  • [EXECUTION_QUALITY]: 95 - 5 points deducted for the source-hygiene issue only; code flow is simple, static checks passed, CI is green, and the targeted unit spec passed locally.
  • [PRODUCTIVITY]: 90 - 10 points deducted because #12130's functional goals are met, but the PR still needs one small cleanup before becoming merge-ready.
  • [IMPACT]: 45 - Localized but meaningful AI-ingestion weighting fix: it prevents external/tenant deployments from turning the community multiplier into a boost-all path.
  • [COMPLEXITY]: 30 - Low complexity: two files, one static registry-derived set, one pure decision helper, and focused tests.
  • [EFFORT_PROFILE]: Quick Win - High value for external-deployment correctness with a small, contained implementation surface.

After that cleanup, this should be an approval-path PR rather than another architecture cycle.

neo-gpt
neo-gpt APPROVED reviewed on May 28, 2026, 9:13 PM

PR Review Follow-Up Summary

Status: Approved

Cycle: Cycle 2 follow-up / re-review

Opening: Re-reviewing the prior CHANGES_REQUESTED state against the new head 5aa69a898, specifically the requested removal of transient ticket anchors from the added IssueIngestor test source.

Strategic-Fit Decision

Per §9 Strategic-Fit Step-Back:

  • Decision: Approve
  • Rationale: The only prior blocker was durable test-source metadata hygiene, and the new delta removes exactly that without changing the behavior under review. Current CI and local targeted validation are green.

Prior Review Anchor

Delta Scope

Summarize what changed since the prior review:

  • Files changed: test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs
  • PR body / close-target changes: unchanged; Resolves #12130 remains the valid close target.
  • Branch freshness / merge state: mergeStateStatus: CLEAN; branch checked out at 5aa69a898.

Previous Required Actions Audit

  • Addressed: Remove the transient #12130 anchors from the added test source at IssueIngestor.spec.mjs:213 and :218 — commit 5aa69a898 changes the block title to Community-multiplier internal-author resolution and the suite title to Neo.ai.daemons.services.IssueIngestor — community-multiplier internal-author set.

Delta Depth Floor

  • Documented delta search: I actively checked the exact diff from 33590a56d to 5aa69a898, scanned the changed source for #12130, #12144, lane, AC, and close-keyword residue, and re-ran the targeted IssueIngestor unit spec; I found no new concerns.

Conditional Audit Delta

Rhetorical-Drift Delta: Pass. The changed test comment and suite title now describe behavior only; no transient ticket/source-cycle anchor remains in the added test surface.

N/A Audits — 📑 📡 🔗

N/A across listed dimensions: the follow-up delta is a test-source wording cleanup only; it does not alter public contracts, MCP tool descriptions, or cross-skill conventions.

Test-Execution & Location Audit

  • Changed surface class: test-source wording only
  • Location check: pass — unchanged canonical path under test/playwright/unit/ai/services/ingestion/
  • Related verification run: git diff --check origin/dev...HEAD passed; node --check ai/services/ingestion/IssueIngestor.mjs passed; node --check test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs passed; npm run test-unit -- test/playwright/unit/ai/services/ingestion/IssueIngestor.spec.mjs -> 7 passed.
  • Findings: pass

Contract Completeness Audit

  • Findings: N/A for this follow-up delta; the functional contract audit from the prior review remains valid and the fixup did not alter implementation behavior.

Metrics Delta

Metrics are unchanged from the prior review unless an explicit delta is listed below.

  • [ARCH_ALIGNMENT]: 95 -> 100 - The only prior deduction was durable test-source ticket-anchor residue; the delta removes it, and I actively checked source-anchor residue, close-target semantics, and the unchanged registry-derived behavior.
  • [CONTENT_COMPLETENESS]: 85 -> 100 - The prior 15-point deduction was specifically for #12130 in the added test block/suite title; commit 5aa69a898 removes those anchors while retaining behavior-focused wording.
  • [EXECUTION_QUALITY]: 95 -> 100 - The source-hygiene issue is resolved, static checks pass, CI is green, and the targeted unit spec passes locally.
  • [PRODUCTIVITY]: 90 -> 100 - The prior required action is addressed fully; the PR now satisfies the review path for #12130 without outstanding merge-blocking work.
  • [IMPACT]: unchanged from prior review at 45 - Localized but meaningful AI-ingestion weighting fix; the follow-up delta does not change impact scope.
  • [COMPLEXITY]: unchanged from prior review at 30 - Low complexity remains accurate: two-file implementation plus focused test coverage, with a tiny wording-only follow-up.
  • [EFFORT_PROFILE]: unchanged from prior review at Quick Win - High value for external-deployment correctness with a small, contained implementation surface.

Required Actions

No required actions — eligible for human merge.

A2A Hand-Off

After posting this follow-up review, I will send the review id and URL to the author via A2A for scoped follow-up.