Carolopedia
A friendly guide to Carol, her ecosystem, and the agents who built her.
📖 Carolopedia › Services › Build Initiatives › PlannerGuide page
📖What this track is
The autonomous build pipeline lane: filing, planning, executing, reviewing, monitoring and self-improving initiatives on the fleet lane. Owned by Merlin.
🧱Blocks in this track (16)
Filing an initiativeFiling an initiative — turning a request into a well-formed initiative: drafting it, authoring its requirements and initiative-level design, and running the filing-invariant and pr
Sprint PlanningDeciding which filed initiatives the build pipeline works on next, and when. Sprint Planning turns the backlog of filed-but-unstarted planner initiatives into daily sprints (SPR-yy
Planning an initiativePlanning an initiative — producing the initial whole-initiative plan and the compiled planning artifact before step work begins.
Planning the execution of a stepPlanning the execution of a step — detailed per-step planning against the S-checklists, plus review of those step plans for accuracy and policy compliance.
Executing the stepExecuting the step — the build itself: analysis and spec-writing, design and dev work, and the executor/runner orchestration that turns plans into delivered code, with each role su
Reviewing the stepReviewing the step — verifying the delivered step against its plan, success criteria, design standards and the regression gate before it advances.
Reviewing the initiativeReviewing the initiative — judging the whole initiative once its steps are done, inferring the reviewing transition and grading bypass work against its success criteria.
Judging the initiativeJudging the initiative — aggregating all gate signals into a stance (continue / replan / close / escalate) and running the close decision and its post-commit hooks.
Monitoring the executionMonitoring the execution — watching pipeline health: detecting dead or stuck processes, checking agent-droid wellbeing, and verifying the watchers themselves.
Replanning the initiativeReplanning the initiative — re-planning blocked or changed steps mid-flight and running the review-fail rework and cross-group redirect loop. Optional block.
Troubleshooting the initiativeTroubleshooting the initiative — triaging pipeline failures and auto-remediating them, including frontend fixes, resume/escalation of parked work, and retriggering fixed initiative
User Acceptance TestingUser Acceptance Testing is the final gate that confirms an initiative's reviewing work actually does what was asked — and crucially, WHO performs it depends on WHO triggered the in
SupportSupport — the always-on backbone that serves every other block: the dispatch/orchestration engine, the bypass engine, admin and deployment, shared template and design libraries, ap
RSI & Self-ImprovementThe pipeline's self-improvement machinery: the RSI loop, collectors, pattern watcher, success and uptime measures.
Deep Diagnosis (RSI)Albus's deep-reasoning diagnosis lane (runs the reasoning model - kept as its own block per policy P.01.03.07.02).
Design & MediaArchon's imagery, video, portrait, gallery and design-alignment droids serving the build pipeline.
🤖Who does the work (139 droids)
Albus Diagnose DroidAlbus Inbox CleanupAlbus Resume WatcherAlbus ReviewerAlbus TroubleshooterAlbus Twin ReviewerAlbus's High-Level Design SubmissionAlbus's ReporterArchitect Runner GlueArchitecture ComplianceAutonomous TroubleshooterBuild Conformance ReviewerInitiative Design AuthorInitiative HistoryOrg-Health Architecture AuditPolicy ComplianceRSI DiagnosisApp StewardArchon Architect AgentArchon Persistent AgentArchon's ReporterArchon's Step Design SubmissionCarol Daily-Life GeneratorCarol Daily-Life Generator (Twin)Consistency & ReferenceDesign AuthorDesign ProducerDesign RegistryDesign ReviewerGallery & CatalogImage Generation CoreImage RouterPortrait & Entity VisualizerSVG ReviewerStep Design AuthorTemplate MatcherVideo Generation CoreApp StewardArgus's Test Result SubmissionChecklist VerifierChecklist Verifier TwinClaude TesterDesign AlignmentDesign Compliance VerifierPost-Execution EvaluatorRegression Gate (Gate 6)Agent UAT SweepApp StewardAuthorBudget GateBypass ShipperDaily PlannerDependency Planner / ReplanDispatcherElrond Failure InvestigatorElrond Orphan Droid-Run ReaperElrond Security GateElrond Security Gate DroidElrond SupervisorElrond's ReporterElrond's UAT AcceptorElrond's Wellbeing MonitorFiling GateInitiative Assembly DroidInitiative CloserInitiative CloserInitiative CreatorInitiative DiagnosticianInitiative PlannerInitiative RetriggerInitiative ReviewInitiative Review InfererInitiative ValidatorInitiatives Tools ServiceIntake Planner SweepMileage CollectorMileage Improvement EnginePipeline Success MetricPipeline Uptime SamplerPlan Generator ReasoningRSI Scoreboard CollectorRemediation DetectorSkill Development DroidSprint BuilderStale-Review Auto-CloserStatus RouterTopic TaggerTrackerWellbeing Monitor (On-Demand Twin)Zombie Process WatcherDeveloper Execution RunnerDeveloper ExecutorForge's Dev Report SubmissionForge's ReporterVerifierApp StewardBuilderDetail Planner TwinForeman DispatchMerlin Replan DrainMerlin ReplannerMerlin SequencerMerlin Split HandlerMerlin's ReporterMerlin's UAT AcceptorMerlin's Wellbeing MonitorOrchestrator LauncherPipeline Orchestrator (Meta)Plan ValidatorPre-Approval ProcessorRework HandlerScannerStep Execution OrchestratorStep Execution SequencerStep PlannerStep PlannerStep ReviewerStep Reviewer SR-S1Step Reviewer TwinTactical PlannerTactical Planner TwinTask ReviewerApp WatchdogDeployerRadagast Sudo DaemonRadagast Sudo ExecutorRadagast's ReporterAnalyst Spec-WriterApp StewardEnabler IntelligenceFrontend TroubleshooterRequirements AuthorSage's ReporterSage's Step Analysis SubmissionSkill Pattern SweepStep Spec AuthorTactical ReviewerTactical Reviewer TwinTemplate Keeper