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.

Neo.mjs - Self-Evolving Software Organism

Frontmatter

title	fix(codex): activate stop hook blocking (#13661)
author	neo-gpt
state	Merged
createdAt	Jun 20, 2026, 10:42 PM
updatedAt	Jun 20, 2026, 10:53 PM
closedAt	Jun 20, 2026, 10:53 PM
mergedAt	Jun 20, 2026, 10:53 PM
branches	`dev` ← `codex/13661-codex-stop-blocking`
url	https://github.com/neomjs/neo/pull/13662

Merged

neo-gpt commented on Jun 20, 2026, 10:42 PM

Resolves #13661

Summary: Activates the Codex Stop hook in enforcement mode instead of dry-run mode. The committed hook config now exports NEO_CODEX_LANE_STATE_ENFORCE=1, the Codex hook reports real decision: "block" output when a turn-terminal message lacks operator-prompt evidence, and the focused Codex hook tests assert the blocking path through the spawned hook.

Related: #13623, #13624, #13655, #13657, #13659, #13660

Evidence: L2 (focused unit coverage + spawned-hook stdout/block-log validation in sandbox) -> L3 required post-merge, because final proof is a restarted Codex harness receiving a Stop event and refusing an invalid autonomous turn-end.

Deltas

None. This is the direct P0 activation lane for Codex Stop-hook blocking; hook-internal read/parse failures remain fail-open and audit-only.

Test Evidence

node --check .codex/hooks/codex-lane-state-stop.mjs
node --check test/playwright/unit/hooks/codexLaneStateStopHook.spec.mjs
npm run test-unit -- test/playwright/unit/hooks/codexLaneStateStopHook.spec.mjs
npm run test-unit -- test/playwright/unit/hooks/codexLaneStateStopHook.spec.mjs test/playwright/unit/hooks/laneStateStopHook.spec.mjs
npm run test-unit -- test/playwright/unit/hooks/codexLaneStateStopHook.spec.mjs test/playwright/unit/hooks/laneStateStopHook.spec.mjs test/playwright/unit/ai/scripts/lifecycle/stopHookDecision.spec.mjs
Manual enforced spawned-hook check:
- NEO_CODEX_LANE_STATE_ENFORCE=1 NEO_AI_DAEMON_DIR=/private/tmp/codex-stop-hook-manual-13661-rebased node .codex/hooks/codex-lane-state-stop.mjs
- Invalid payload emitted stdout {"decision":"block",...} and wrote BLOCK to the Codex Stop-hook audit log.

Post-Merge Validation

Restart Codex from dev, trigger an invalid autonomous Stop, and confirm the hook blocks the turn instead of accepting dry-run termination.

Commit:

0e77b9010 - fix(codex): activate stop hook blocking (#13661)

Authored by Euclid (GPT-5 Codex, Codex Desktop). Session 810318e6-b644-474e-a255-a07d19825aa5.

github-actions commented on Jun 20, 2026, 10:42 PM

🚨 Agent PR Body Lint Violation

@neo-gpt — your PR body on PR #13662 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 ## Test Evidence is missing.

Visible anchors missing (full list)

## Test Evidence
## 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.}

tobiu APPROVED reviewed on Jun 20, 2026, 10:53 PM

No review body provided.