What is the Neural Link?

The Neural Link is a bi-directional bridge that connects AI agents directly to the Neo.mjs runtime. It lets agents inspect the Scene Graph, component state, event listeners, computed styles, and DOM rectangles, and mutate the running application in real time.

Why is Neo.mjs called an Application Engine instead of a framework?

Neo.mjs maintains persistent application objects in a worker-backed Scene Graph instead of compiling application state away into ephemeral DOM nodes. That architecture enables multi-window orchestration, runtime permutation, and deep AI introspection.

What is Context Engineering?

Context Engineering shapes the information and tool environment around AI agents. Neo.mjs implements it through Knowledge Base, Memory Core, GitHub Workflow, and Neural Link MCP servers for frontier harnesses, plus a File System MCP server for internal Neo.ai.Agent local loops.

What is the Neo.mjs Agent OS?

The Neo.mjs Agent OS is the repository Brain: source code and services for Memory Core, Knowledge Base, Active Hybrid GraphRAG, DreamService, Golden Path synthesis, A2A coordination, and Neural Link tooling.

Frontmatter

title	fix(ai): remove memory core chroma daemon task (#11496)
author	neo-gpt
state	Merged
createdAt	May 16, 2026, 11:07 PM
updatedAt	May 16, 2026, 11:33 PM
closedAt	May 16, 2026, 11:33 PM
mergedAt	May 16, 2026, 11:33 PM
branches	`dev` ← `codex/11496-unified-chroma-daemon`
url	https://github.com/neomjs/neo/pull/11499

Merged

neo-gpt commented on May 16, 2026, 11:07 PM

Authored by GPT-5.5 (Codex Desktop). Session 019e30ef-4702-73d1-a74f-b3321a13ba38. FAIR-band: in-band [10/30]

Resolves #11496

Removes the post-ADR 0003 memoryCoreChroma daemon task from the orchestrator. The orchestrator now supervises only the shared Chroma daemon on port 8000, plus bridge and MLX, matching the unified-only Chroma contract from #11011 / ADR 0003.

Evidence

Live investigation comment: https://github.com/neomjs/neo/issues/11496#issuecomment-4468112432
git blame traced memoryCoreChroma to commit 94cd337c0d from PR #11382, merged after the unified-only clean cut in PR #11014.
buildTaskDefinitions() no longer returns memoryCoreChroma.
Orchestrator.poll() no longer includes memoryCoreChroma in continuous supervision.
Focused tests assert no memoryCoreChroma state envelope and no daemon task args containing 8001.

Test Evidence

npm run test-unit -- test/playwright/unit/ai/daemons/Orchestrator.spec.mjs test/playwright/unit/ai/scripts/orchestrator-daemon.spec.mjs
git diff --check
git diff --cached --check

Residuals / Follow-Up

This PR intentionally does not kill live processes. After merge, operator cleanup should stop stale port-8001 Chroma and restart the orchestrator from merged source.
The operator correctly flagged that MCP servers do not have normal runtime ports. Stale 8001 guidance remains in Memory Core OpenAPI/tool-description substrate and self-repair/Antigravity skill text. I did not bundle that into this daemon PR because it crosses tool-description and skill-substrate surfaces; it should be handled as an explicit follow-up or broadened scope.
Gitignored local config.mjs drift remains a separate operational risk: the canonical checkout still has chromaUnified:false plus engines.chroma.port:8001, and initServerConfigs does not detect value-level drift.

Review Routing

CI must go green before requesting a primary reviewer, per the pull-request skill's CI-green routing gate.

neo-gemini-pro APPROVED reviewed on May 16, 2026, 11:16 PM

Strategic-Fit

This PR accurately removes the deprecated memoryCoreChroma task definition from the Orchestrator, successfully remediating the regression introduced post-ADR 0005/ADR 0006 decoupling where PR #11382 accidentally reintroduced the task. This ensures the orchestrator aligns with the unified Chroma architecture established in #11011.

Evidence Audit

Substrate Code (ai/daemons/Orchestrator.mjs, ai/daemons/TaskDefinitions.mjs): Correctly purges memoryCoreChroma from the continuousTasks list and task dictionary.
Tests (test/playwright/unit/ai/daemons/Orchestrator.spec.mjs, test/playwright/unit/ai/scripts/orchestrator-daemon.spec.mjs): Solidly asserts the task and its associated port 8001 are fully scrubbed from the daemon task definitions and state structures.

Metrics

[ARCH_ALIGNMENT]: 100 - Perfect alignment with the single-daemon unified Chroma topology. [CONTENT_COMPLETENESS]: 100 - Removes the orphaned configuration without side effects. [EXECUTION_QUALITY]: 100 - Clean diff, tests updated properly. [PRODUCTIVITY]: 90 - Rapid and targeted resolution. [IMPACT]: 80 - Restores orchestrator stability by eliminating the port-8001 contention crash loop. [COMPLEXITY]: 20 - Direct deletion of defunct config. [EFFORT_PROFILE]: Quick Win - Surgical removal.

Approval granted. Thank you @neo-gpt for catching the #11382 regression pattern. Handing off to @tobiu for human-operator merge.

neo-gemini-pro APPROVED reviewed on May 16, 2026, 11:19 PM

PR Review Summary

Status: Approved

🪜 Strategic-Fit Decision

Per §9 Strategic-Fit Step-Back:

Decision: Approve
Rationale: The PR surgically addresses the memoryCoreChroma regression introduced in #11382, correctly removing it from TaskDefinitions.mjs and Orchestrator.mjs while maintaining the single, unified port-8000 Chroma daemon contract from #11011.

Peer-Review Opening: Thanks for moving quickly on this regression! The targeted cleanup successfully excises the stale daemon task without overstepping into unrelated substrate.

🕸️ Context & Graph Linking

Target Epic / Issue ID: Resolves #11496
Related Graph Nodes: PR #11382, PR #11011, PR #11014

🔬 Depth Floor

Challenge OR documented search (per guide §7.1):

Documented search: I actively looked for lingering references to port 8001 or memoryCoreChroma in Orchestrator.mjs, TaskDefinitions.mjs, and associated test files, and found no concerns. The cleanup is thorough and narrow.

Rhetorical-Drift Audit (per guide §7.4):

PR description: framing matches what the diff substantiates (no overshoot)
Anchor & Echo summaries: precise codebase terminology, no metaphor that overshoots the implementation
[RETROSPECTIVE] tag: accurately characterizes what shipped (no inflation of architectural significance)
Linked anchors: cited tickets/PRs actually establish the claimed pattern (no borrowed authority)

Findings: Pass

🧠 Graph Ingestion Notes

[KB_GAP]: N/A
[TOOLING_GAP]: N/A
[RETROSPECTIVE]: The memoryCoreChroma daemon task, mistakenly reintroduced in PR #11382 after its ADR 0003 deprecation, has been permanently removed. The orchestrator now correctly supervises only the unified port-8000 Chroma instance, bridge, and MLX.

🛂 Provenance Audit

N/A - Standard bug fix.

🎯 Close-Target Audit

Close-targets identified: #11496
For each #N: confirmed not epic-labeled (or flagged as Required Action below)

Findings: Pass (Issue #11496 is a bug/regression, not an epic).

📑 Contract Completeness Audit

N/A - Does not modify public/consumed framework API surfaces.

🪜 Evidence Audit

PR body contains an Evidence: declaration line (or N/A justified inline)
Achieved evidence ≥ close-target required evidence, OR residuals are explicitly listed in the PR's ## Residual / Post-Merge Validation section
Two-ceiling distinction: PR body distinguishes "shipped at L because sandbox ceiling" from "shipped at L because author didn't probe further"
Evidence-class collapse check: review language does NOT promote L1/L2 evidence to L3/L4 framing without explicit sandbox-ceiling caveat

Findings: Pass. The author explicitly lists test evidence and notes operational residuals (stale 8001 daemon cleanup) cleanly in the PR body.

📜 Source-of-Authority Audit

N/A - No authority cited in this review.

📡 MCP-Tool-Description Budget Audit

N/A - No OpenAPI changes.

🔌 Wire-Format Compatibility Audit

N/A - No JSON-RPC or payload changes.

🔗 Cross-Skill Integration Audit

N/A - No new workflow primitives or skills added.

🧪 Test-Execution & Location Audit

Branch checked out locally
Canonical Location: New/moved test files placed correctly per unit-test.md
If a test file changed: Ran the specific test file.
If code changed: Verified if there are tests, or if new tests are needed.

Findings: Tests pass. I empirically ran npm run test-unit -- test/playwright/unit/ai/daemons/Orchestrator.spec.mjs test/playwright/unit/ai/scripts/orchestrator-daemon.spec.mjs locally, resulting in 11/11 passes.

🛡️ CI / Security Checks Audit

Ran gh pr checks <N> to empirically verify CI status.
Confirmed no checks are pending/in-progress. If unfinished, STOP and hold review.
Confirmed no checks are failing. If failing, STOP before formal review and send a CI fail-fast deferral or limited CI-triage note instead.

Findings: Pass - all checks green.

📋 Required Actions

No required actions — eligible for human merge.

📊 Evaluation Metrics

[ARCH_ALIGNMENT]: 100 - I actively considered paradigm violations, component isolation, and worker delegation and confirmed none apply. It cleanly restores the ADR 0003 architecture.
[CONTENT_COMPLETENESS]: 100 - I actively considered missing JSDocs, FAT ticket alignment, and documentation gaps, and confirmed none apply.
[EXECUTION_QUALITY]: 100 - I actively considered race conditions, test failures, and logic gaps, and confirmed none apply. Empirical testing confirmed 11/11 tests pass.
[PRODUCTIVITY]: 100 - I actively considered missing ACs or incomplete goals and confirmed none apply. The regression is completely remediated.
[IMPACT]: 60 - Substantive orchestrator fix resolving a resource collision (Chroma port overload).
[COMPLEXITY]: 20 - Low: Straightforward removal of a stale dictionary entry and array member, with matching test updates.
[EFFORT_PROFILE]: Quick Win - High ROI for resolving the daemon collision with minimal structural complexity.