Backlog

Tickets 1701-1742
Tickets 1601-1701
Tickets 1501-1599
Tickets 1401-1501
Tickets 1301-1399
Tickets 1201-1300
Tickets 1101-1200
Tickets 1001-1101

11218 Implement 4-Tier Decision Escalation Ladder to Eliminate "Helpful Assistant" Regression Drift
11217 Consensus mandate for ideation-sandbox graduation + PR-merge gate
11213 Session Sunset Protocol: Agent-Level vs Swarm-Level Split
11211 Session Sunset Protocol: Agent-Level vs Swarm-Level Split
11209 Graduate Discussion #11206 Option A-prime — refined lead/peer-role coordination protocol
11205 lead-role-mode.md: mandate explicit /peer-role skill-trigger naming in A2A messages
11204 Harden test harness mocking for dynamic ESM imports
11202 Backfill hasCoreSwarmParticipant misses 33 sessions with non-shared userId
11201 Fix HealthService backup.error: fs.readdir is not a function
11198 IssueService archive-write path refactor for lazy ordinal chunking (#11187 AC3)
11197 DiscussionService flat active-tier and lazy ordinal archive refactor (#11187 AC5)
11196 PullRequestService archive-write path refactor for lazy ordinal chunking (#11187 AC4)
11195 Validate Step 2.5 Architectural Step-Back post-merge (30-day observation)
11192 Graduate Step 2.5 Architectural Step-Back from Discussion #11188
11190 Add archivePath helper for sealed archive chunks
11189 Phase 1: Refactor Config for Archive Shape Management
11187 Adopt single-root archive with lazy 100-item chunking (Discussion #11180 graduation)
11185 Detect stale magic-close keywords in PR commit bodies
11182 Wake subscription duplicate-accumulation + GPT zero-delivery (post-#10624 recurrence)
11181 Restore shared summary visibility for swarm identities
11179 Implement Boot-Time Pre-Flight and Hard-Refusal Predicate for Sandman Handoff
11177 Persist active-PR-cycle-state in sandman_handoff for nightshift boots
11169 Add operator-local config for orchestrator dev-sync roots
11165 post-review-pickup §4 criterion #1 condones deference-slip — revise to mandate backlog self-select + announce
11163 bootstrapWorktree: generate dist/parse5.mjs (+ minimal build artifacts) so worktree can run full unit suite
11161 Migrate knowledge-base services to SDK (M6 Tier-1)
11160 Lead-role end-of-turn Pre-Flight reasoning-statement + anti-deference anti-patterns
11156 Generalize Worktree Bootstrap Protocol to Antigravity
11155 Restore github-workflow projectRoot to v12.1.0 cwd-primary shape (revert #11149)
11154 Extend §7.4 Rhetorical-Drift Audit to cross-PR reviewer-seeded drift
11152 Hardening GitHub Workflow pathing and documentation
11150 Harden production-scale ai:restore path: Neo bootstrap + chunked Chroma upsert + streaming validateBundle
11148 Reject sync_all MCP tool when caller's working tree is not on dev branch
11147 Fix github-workflow MCP config projectRoot pathing resolution
11145 Reject sync_all MCP tool when caller's working tree is not on dev branch
11144 ai:restore Chroma preserve-live parity for #importMemories (follow-up to #11141)
11141 Enhance ai:restore merge mode: preserve-live INSERT OR IGNORE semantics + label-filter + per-substrate targeting + post-restore hooks
11140 Patch Store.clear() cascade bypass of destructive operation guards in SQLite.removeNodes
11138 fix(github-workflow): LocalFileService.getIssueById missed chunkPath in active-issue read path
11136 A2A peer-engagement: name skills literally in routing requests (e.g., /pr-review on PR #N)
11135 Configure dev auto-sync for multiple repo clones
11134 enhancement: mitigate systemic friction from interleaving chore-sync commits on active branches
11133 GH-content sync: workflow-fragility — local branch-discipline gap producing stale-branch + chore-sync contamination (5x same-session pattern)
11131 Integrate PR sync script into release publish workflow
11128 peer-role-mode.md: codify pro-active essential + session-sunset duration (sister to #11124)
11126 feature(github-workflow): Implement write-path chunking for IssueSyncer
11124 lead-role-mode.md: codify session-sunset duration + positive-framed lane-coordination pattern
11121 Migrate PR conversations data to hierarchical chunking (XXxx)
11118 Chunk resources/content/discussions/ to bypass GitHub's 1000-file folder cap (preventive sibling to #11113)
11117 Chunk resources/content/pulls/ to bypass GitHub's 1000-file folder cap (sibling to #11113)
11116 Epic-shape discipline: separate code-change commits from data-migration commits in PRs
11115 Institutionalize Safeguards Against Chore-Sync Contamination in Feature Branches
11113 Subdivide issue directories to bypass GitHub's 1000-item folder limit
11111 ChromaManager dedup: extract shared Chroma-client primitive (M6 retrospective)
11110 github-workflow MCP needs systematic boot + wiring + integration test coverage
11108 Remove window duration telemetry from A2A wake messages
11107 Move toolService.mjs files back to MCP server directories (M6 retrospective architectural fix)
11105 pull-request skill: add author-side check that reviewers use correct pr-review template
11103 github-workflow MCP boot fails: toolService.mjs has stale 3-dot relative import (post-#10997 M6 migration)
11102 epic-resolution Verdict-Authority calibration: peer-V-B-A consensus suffices for epic close
11099 Compress Double Diamond skill rules into Atlas payloads
11094 Event-Driven Issue Sync and Golden Path Synthesis
11093 Migrate Daemon Supervision to AgentOrchestrator
11092 AGENTS.md core-value triggers: friction→gold §13.X + V-B-A §3.5
11090 Compress new skill rules into Atlas payloads
11086 Add Double Diamond graduation guard to ideation-sandbox
11084 pr-review skill: Cycle-1 premise pre-flight + Drop+Supersede framing triggers
11082 Implement Double Diamond guard in Ideation Sandbox
11081 Rename Orchestrator.mjs to AgentOrchestrator.mjs
11080 Orchestrator daemon fails to start (Sub-4 #11064 regression)
11078 Enforce Double Diamond divergence guard in ideation-sandbox
11077 M4 Architectural Convergence: Orchestrator Supervision & Unified Daemons
11075 Exploration: Migrate orchestrator magic numbers + retention constants to ai/config.template.mjs
11074 Extract WakeSubstrateCoordinatorService as M4 per-task coordinator (fold bridge-daemon into orchestrator)
11073 Extract HeartbeatCoordinatorService as M4 per-task coordinator (fold SwarmHeartbeatService into orchestrator)
11072 Extract GraphMaintenanceCoordinatorService as M4 per-task coordinator (Hebbian decay + GC scheduling)
11071 Extract GoldenPathCoordinatorService as M4 per-task coordinator (Golden Path synthesis scheduling)
11070 Extract DreamCoordinatorService as M4 per-task coordinator (DreamService cycle scheduling)
11068 M3.5 Sub-4: Orchestrator slim-down + TaskDefinitions extraction
11066 bridge-daemon launchd/systemd packaging (close operator's bridge terminal)
11065 Extract SandmanCoordinatorService as M4 per-task coordinator (orchestrator-driven Sandman/DreamService scheduling)
11062 Extract BackupCoordinatorService as M4 per-task coordinator (orchestrator daily backup)
11058 Split SwarmHeartbeatService into class file + ai/scripts/ entry-point wrapper (Orchestrator pattern)
11057 Abstract fixture identities in team/private retrieval spec
11052 [ai] Update lead/peer role SKILL.md to reflect session-sunset exit conditions
11051 [ai] Extract CadenceEngine from Orchestrator (M3.5 Sub-3)
11050 [ai] Refine Lead/Peer Role Exit Conditions to Align with Session Sunset
11049 ai/ Neo namespace bootstrap hygiene: entry-point-only Neo+core+InstanceManager imports
11046 [Bug] Verify ai:syncKnowledgeBase.mjs behavior
11039 Extract TaskStateService to isolate state-mutation API
11038 Implement Autonomous Swarm Lead Rotation (A2A Baton Pass)
11032 [Skill Enhancement] Enforce assignee check in pull-request Pre-Flight to mitigate 30% unassigned closure rate
11031 Create /peer-role skill — SKILL.md + reference payload + AGENTS.md §21 row (graduates Discussion #11026)
11030 Add AGENTS.md §15.6 Swarm Topology Anchor — Flat Peer-Team Model (graduates Discussion #11026 OQ4)
11029 Add per-recipient read receipts for A2A broadcasts
11028 Create /lead-role skill — SKILL.md + reference payload + AGENTS.md §21 row (graduates Discussion #11024)
11027 Codify "contributions over commits" in AGENTS.md §13 + Atlas qualitative measurement framework
11022 Orchestrator decomposition (M3.5): extract ProcessSupervisorService, TaskStateService, CadenceEngine
11020 gh-workflow sync: skip NOT_PLANNED snapshots + on-release archival automation
11019 v13-path.md substrate update: D3 scope additions for backup, FS ingestor, cadence design, MX metrics
11017 Orchestrator task: primary-checkout `dev` auto-sync + KB cascade

Tickets 901-1000
Tickets 801-901
Tickets 701-800
Tickets 601-699
Tickets 501-599
Tickets 401-500
Tickets 301-400
Tickets 201-300
Tickets 101-200
Tickets 1-100

v12.1.0

Tickets 201-213
Tickets 101-200
Tickets 1-100

Frontmatter

id	11213
title	Session Sunset Protocol: Agent-Level vs Swarm-Level Split
state	Closed
labels	enhancementai
assignees	[]
createdAt	May 11, 2026, 4:28 PM
updatedAt	May 11, 2026, 6:18 PM
githubUrl	https://github.com/neomjs/neo/issues/11213
author	neo-gemini-pro
commentsCount	1
parentIssue	null
subIssues	[]
subIssuesCompleted	0
subIssuesTotal	0
blockedBy	[]
blocking	[]
closedAt	May 11, 2026, 6:18 PM

Session Sunset Protocol: Agent-Level vs Swarm-Level Split

Closed Backlog/active-chunk-11 enhancementai

neo-gemini-pro commented on May 11, 2026, 4:28 PM

Refines the sunset protocol by introducing explicit scope declarations (solo-refresh vs convergent) to prevent token-refresh sunsets from triggering global swarm-level halts. Adds path-determinism rules (IF/ELSE-IF) and requires explicit scope declarations for the next 3 sunsets as a validation hook.

tobiu referenced in commit f2f8fd1 - "feat(agents): implement sunset scope split (solo-refresh vs convergent) (#11213) (#11215) on May 11, 2026, 6:18 PM

tobiu closed this issue on May 11, 2026, 6:18 PM