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

Latest
v13.1.0

PRs 701-717
PRs 601-699
PRs 501-600
PRs 401-500
PRs 301-400
PRs 201-300
PRs 101-200
PRs 1-130

v13.0.0

PRs 1301-1325
PRs 1201-1300
PRs 1101-1200
PRs 1001-1100
PRs 901-1000
PRs 801-900
PRs 701-800
PRs 601-700
PRs 501-600

11276 feat(github-workflow): atomic PR review create/update via dedicated MCP tool (#11273)
11272 chore(deps-dev): bump @protobufjs/utf8 from 1.1.0 to 1.1.1
11271 fix(agentos): repair pre-flight skill discovery (#11270)
11269 chore(agentos): shorten 3 over-floor SKILL.md routers via Map→Atlas relocation (#11268)
11266 feat(agent): implement sandman handoff pre-flight skill (#11260)
11264 feat(agentos): introduce /sandman-handoff-pre-flight skill substrate (#11260)
11263 chore(agents): apply XML tags for anti-drift anchors (#11262)
11261 refactor(agent): relocate sandman handoff to dedicated skill (#11260)
11258 docs(pr-review): add loading-runtime-effect substitution anti-pattern (#11256)
11257 feat(agent): implement boot-time sandman handoff pre-flight guard (#11179)
11255 feat(agentos): institutionalize architectural pre-flight skills (#11256)
11251 chore: remove obsolete migrateV13ProjectFromLabel script (#11233)
11250 docs(peer-role): add loading-runtime-effect substitution anti-pattern (#11248)
11249 docs(pr-review): add source-ticket currency preflight (#11235)
11247 chore: remove obsolete reconcileV13Project script (#11233)
11246 chore(ai/scripts): Add Phase 2 ProjectV2 migration script (#11233)
11245 docs(peer-role): integrate schlagfertig-discipline and discipline-dressed-deference (#11243)
11244 feat(ai): implement 3-layered defense — L1 firewall + L2 premise-risk + L3 reflective pause (#11241)
11242 feat(ai): implement layer 2 premise-risk guardrails and reflective pause (#11241)
11234 feat(github-workflow): extend create_issue with projects param + add manage_issue_projects tool (#11233)
11232 refactor(mcp): move 4 toolService.mjs files back to MCP server dirs (#11107)
11231 chore(deps-dev): bump mermaid from 11.12.3 to 11.15.0
11230 fix(skills): correct epic-resolution-workflow §4 verdict-authority misframing (#11229)
11228 feat(mc): enforce Zod-validated SDK perimeter (#11224)
11227 fix(github-workflow): paginate ProjectV2 items in reconcile script to bypass 100-item limit (#10961)
11226 test(ai): guard dynamic fs-extra import shape (#11204)
11225 refactor(memory-core): migrate services to SDK flatter structure (#11224)
11223 feat(agents): graduate Discussion #11206 Option A-prime — lead/peer coordination protocol (#11209)
11222 fix(agents): close §15.6 OR loophole — announce AND execute (#11221)
11220 feat(agents): implement 4-Tier Decision Escalation Ladder (#11218)
11219 feat(agents): consensus mandate for ideation-sandbox graduation + PR-merge gate (#11217)
11215 feat(agents): split sunset scope into agent and swarm levels
11212 feat(agents): split sunset scope and codify uncertainty rule
11208 feat(agents): mandate explicit /peer-role skill-trigger naming in lead-role A2As (#11205)
11207 fix(scripts): resolve migration tag naming collision (#11202)
11203 fix(memory-core): pass fs-extra default export to buildBackupStateBlock (#11201)
11200 fix(memory-core): share core swarm summaries (#11181)
11199 refactor(github-workflow): use archivePath for pull requests and discussions (#11187)
11194 feat(agents): graduate Step 2.5 Architectural Step-Back (#11192)
11193 feat(github-workflow): add archive path helper (#11190)
11191 feat(github-workflow): introduce unified archiveRoot configuration (#11189)
11186 docs(ai): detect stale close keywords in commit bodies (#11185)
11184 fix(memory-core): reconcile duplicate WAKE_SUBSCRIPTION per route-tuple at bootstrap (#11182)
11183 fix(memory-core): reconcile duplicate WAKE_SUBSCRIPTION at bootstrap (#11182)
11178 feat(ai): persist active PR cycle state in sandman_handoff for nightshift boots (#11177)
11176 docs(ai): mandate backlog self-select to replace deference-slip (#11165)
11175 feat(ai): load local orchestrator dev-sync roots config (#11169)
11174 refactor(ai): migrate discussion files to hierarchical chunking (#11122)
11173 fix(test): restore singleton state in DreamService.spec afterAll + getContextFrontier patch (#10946)
11172 docs(skill): mandate backlog self-select in post-review-pickup (#11165)
11170 docs(worktree): document #11163 build-all scope decision in bootstrapWorktree (#11163)
11167 feat(swarm): mandate backlog self-survey for legitimate halt state (#11165)
11166 feat(skills): codify cross-PR reviewer-seeded drift sub-section in pr-review §7.4 (#11154)
11164 feat(swarm,sync): codify Helpful-Assistant negative constraint in §15.6 (#11128) + chunked-content data migration (#11121)
11162 feat(orchestrator): configure dev sync roots (#11135)
11159 test(memory-core): await GraphService promise wrappers (#10941)
11158 fix(github-workflow): revert 3-tier projectRoot heuristic to cwd-primary shape (#11155)
11157 feat(worktree): finalize bootstrap protocols and antigravity support (#11156)
11153 feat(github-workflow): pnpm workspace layout support and NEO_WORKSPACE_ROOT docs (#11152)
11151 fix(ai-restore): harden production-scale restore path after May 10 recovery (#11150)
11149 fix(ai): resolve projectRoot pathing bug in github-workflow config (#11147)
11146 feat(github-workflow): mechanically REJECT sync_all when caller not on dev branch (#11145)
11143 feat(ai-restore): preserve-live merge semantics + per-incident filter/targeting/hooks (#11141)
11142 fix(graph): inject Database context to Store and guard clear() for production SQLite DB (#11140)
11139 fix(github-workflow): apply chunkPath in LocalFileService active-issue read path (#11138)
11137 docs(skills): name skills literally in A2A peer-routing requests (#11136)
11132 build(release): integrate PR sync into publish workflow (#11131)
11130 feat(ai): add primary dev sync orchestrator task (#11017)
11129 refactor(github-workflow): unify Issue, PR, and Discussion syncers to use chunkPath (#11126)
11127 docs(skills): codify lead-role essential + session-sunset duration + coordination pattern (#11124)
11125 feature(github-workflow): Sync discussions into chunked directories (#11118)
11123 refactor(github-workflow): implement chunked storage for pull requests
11114 feat(github-workflow): implement chunk-based nested folder structure for issues
11109 refactor(memory): remove windowDuration from a2a wake telemetry (#11108)
11106 feat(skill): add author-side reviewer template adherence check (#11105)
11104 fix(github-workflow): correct stale ToolService import path post-M6 migration (#11103)
11101 feat(daemons): migrate daemon supervision to event-driven model (#11094)
11100 docs(skills): compress Double Diamond guard docs (#11099)
11098 docs(agents): elevate friction→gold + verify-before-assert as core values (#11092)
11097 docs(skills): compress pr-review premise pre-flight (#11090)
11096 feat(ai): Centralize daemon supervision (#11093)
11095 feat(skills): Double Diamond graduation guard for ideation-sandbox (#11086)
11088 refactor(agent): rename Orchestrator to AgentOrchestrator (#11081)
11087 fix(ai): restore missing path import and default script resolution in Orchestrator daemon (#11080)
11085 feat(skills): pr-review Cycle-1 premise pre-flight + Drop+Supersede framing (#11084)
11083 enhancement(docs): implement Double Diamond guard in ideation sandbox (#11082)
11069 feat(ai): extract BackupCoordinatorService as M4 per-task coordinator (#11062)
11067 feat(ai): bridge-daemon launchd/systemd packaging (#11066)
11064 refactor(ai): decompose orchestrator into CadenceEngine and extract definitions (#11022)
11063 fix(memory-core): stabilize broadcast sentinel permission spec (#10937)
11061 refactor(ai): split SwarmHeartbeatService into class + entry-point wrapper (#11058)
11060 refactor(test): abstract fixture identities in team/private retrieval spec (#11057)
11059 refactor(ai): extract CadenceEngine from Orchestrator (#11051)
11056 docs(skills): apply Option B clarifier to lead-role and peer-role exit conditions (#11050)
11055 feat(ai): add team-private retrieval integration proof (#10951)
11054 refactor(ai): unify v13 daemon entry-point Neo bootstrap (#11049)
11053 fix(ai): enforce Neo bootstrap hygiene for daemon entry points (#11049)
11048 feat(ai): create peer-role skill (#11031)
11047 fix(ai): correct import paths in syncKnowledgeBase script (#11046)
11045 feat(agentos): document lead-role baton pass (#11038)

PRs 401-500
PRs 301-400
PRs 201-300
PRs 101-242
PRs 1-205

v12.1.0
v12.0.0
v11.24.0
v11.23.0
v11.21.0
v11.18.0
v11.17.0
v11.15.0
v11.14.0
v11.12.0
v11.5.0
v11.4.0
v11.2.0
v11.1.0
v11.0.1
v11.0.0
v10.6.0
v10.5.4
v10.0.0-beta.6
v10.0.0-beta.5
v10.0.0-beta.4
v9.0.0
v8.41.1
v8.41.0
v8.14.0
v8.11.0
v8.1.0

Frontmatter

title	feat(agents): split sunset scope and codify uncertainty rule
author	neo-gemini-pro
state	Closed
createdAt	May 11, 2026, 4:26 PM
updatedAt	May 22, 2026, 8:27 PM
closedAt	May 11, 2026, 4:27 PM
mergedAt
branches	`dev` ← `agent/11211-sunset-scope-split`
url	https://github.com/neomjs/neo/pull/11212

Closed

neo-gemini-pro commented on May 11, 2026, 4:26 PM

Refines the sunset protocol by introducing explicit scope declarations (agent vs swarm) to prevent token-refresh sunsets from triggering global swarm-level halts.

Also codifies the 'Chocolate or Vanilla' rule in AGENTS.md to mandate proactive architectural decision-making via ideation-sandbox instead of passive permission-seeking.

Fixes #11211

neo-gemini-pro commented on May 11, 2026, 4:27 PM

Premature PR. Ideation Sandbox #11210 did not reach graduation state. Closing.