# AlphaSwarm > AgenticOps + RL-Ops platform: strategy research, backtests, paper, and live trading. ## docs ### architecture #### decisions - [ADR 006: alphaswarm_admin overhaul (multi-cloud control plane)](/docs/architecture/decisions/aqp-admin-overhaul.md): - Status: Proposed - [ADR 003 — Auth0 zero-trust two-layer security model](/docs/architecture/decisions/auth0-zero-trust.md): - Status: Accepted (2026-05-18) - [ADR 011 — CDN-fronted standalone container for the cloud-hosted alphaswarm_ui](/docs/architecture/decisions/cdn-fronted-standalone-for-aqp-ui.md): - Status: Accepted (2026-05-25) - [ADR-013: Entra ID as the AlphaSwarm staff first user pool](/docs/architecture/decisions/entra-as-first-pool.md): - Status: Accepted - [ADR-014: Knowledge-Base Boundary](/docs/architecture/decisions/knowledge-base-boundary.md): Status: accepted (2026-05-28) - [ADR 004 — Abstract InfrastructureProvider ABC for workload runtime ops](/docs/architecture/decisions/provider-abstraction.md): - Status: Accepted (2026-05-18) - [ADR 010 — Canary Rollout PnL Gates](/docs/architecture/decisions/quantbot-canary-pnl-gates.md): Status: Accepted (QuantBot Platform v0.2.0) - [ADR 008 — Bot Event Sourcing (PostgreSQL, monthly-partitioned)](/docs/architecture/decisions/quantbot-event-sourcing.md): Status: Accepted (QuantBot Platform v0.2.0) - [ADR 007 — QuantBot Latency Classes](/docs/architecture/decisions/quantbot-latency-classes.md): Status: Accepted (QuantBot Platform v0.2.0) - [ADR 006 — QuantBot Operator Pattern (kopf + Pydantic mirrors)](/docs/architecture/decisions/quantbot-operator-pattern.md): Status: Accepted (QuantBot Platform v0.2.0) - [ADR 009 — MiFID II RTS 6 + SEC 15c3-5 Conformance](/docs/architecture/decisions/quantbot-rts6-conformance.md): Status: Accepted (QuantBot Platform v0.2.0) - [ADR-010: alphaswarm_rl production-grade enhancement (Phases 1-12)](/docs/architecture/decisions/rl-production-enhancement.md): Status: accepted (2026-05-24) - [ADR 015 — Runtime decomposition: cell-based modular monolith over domain microservices](/docs/architecture/decisions/runtime-decomposition.md): - Status: Accepted (2026-06-10) — operator approval recorded during the KG-platform execution run (P0 gate review, T01-T03 complete; Track C repositories now exist and are scaffolded) - [ADR 005 — Separated alphaswarm_controller/ micro-project](/docs/architecture/decisions/separated-control-plane.md): - Status: Accepted (2026-05-18) - [ADR 002 — Single multi-stage container for the AlphaSwarm client surface](/docs/architecture/decisions/single-container-client.md): - Status: Accepted (2026-05-18) — Superseded for alphaswarm_ui by ADR 011 on 2026-05-25. Still in force for the local-operator alphaswarm_client/ packaging path. - [ADR 012 — Solara deprecation in the cloud build](/docs/architecture/decisions/solara-deprecation.md): - Status: Accepted (2026-05-25) - [ADR 001 — Static export (Vite) over SSR for the AlphaSwarm client surface](/docs/architecture/decisions/static-export-over-ssr.md): - Status: Accepted (2026-05-18) #### preprocessing-spec A PreprocessingSpec is a tiny dataclass that travels with every trained - [PreprocessingSpec](/docs/architecture/preprocessing-spec.md): A PreprocessingSpec is a tiny dataclass that travels with every trained ### compliance #### soc2-evidence-map Mapping from the SOC 2 Trust Services Criteria to the - [SOC 2 Type II evidence map](/docs/compliance/soc2-evidence-map.md): Mapping from the SOC 2 Trust Services Criteria to the ### concepts #### agentic - [Agentic development for AlphaSwarm](/docs/concepts/agentic/agentic-development.md): The single doc that connects AlphaSwarm's existing primitives to the - [Agentic pipeline](/docs/concepts/agentic/agentic-pipeline.md): Doc map: intro · - [Agents](/docs/concepts/agentic/agents.md): This document covers the spec-driven agent surface added by the - [Alpha Researcher agent + symbolic alpha DSL](/docs/concepts/agentic/alpha-researcher-agent.md): Self-evolving LLM-driven factor mining wired into AlphaSwarm's - [Bots](/docs/concepts/agentic/bots.md): The smallest self-contained, deployable unit on AlphaSwarm. - [Multi-agent patterns in AlphaSwarm](/docs/concepts/agentic/multi-agent-patterns.md): Catalogue of multi-agent topologies, mapped to existing code in - [Orchestration control plane refactor — rollout runbook](/docs/concepts/agentic/orchestration-refactor-rollout.md): This is the operator-facing rollback / rollout guide for the additive - [Research Agents](/docs/concepts/agentic/research-agents.md): | Spec | Module | Purpose | - [Selection Agents](/docs/concepts/agentic/selection-agents.md): The Selection team picks the top-N tickers for a - [Trader Agents](/docs/concepts/agentic/trader-agents.md): The spec-driven trader (trader.signal_emitter) coexists with the - [Workflow Studio](/docs/concepts/agentic/workflow-studio.md): The Workflow Studio is the operator-facing surface for the additive #### data - [Bi-temporal PermissionedDataPoint](/docs/concepts/data/bi-temporal-graph.md): Every node and every edge in the KB carries the same envelope: - [alphaswarm_kb_federation](/docs/concepts/data/kb-federation.md): The federation gateway is a standalone FastAPI service that brokers - [KB permissions](/docs/concepts/data/kb-permissions.md): Hybrid stack - [KBRuntime + KBCorpusSpec](/docs/concepts/data/kb-runtime.md): Hash-locked spec - [Silo-per-tenant IaC](/docs/concepts/data/kb-silo-iac.md): Section G of the alphaswarm_kb blueprint implemented as a Terragrunt - [AlphaSwarm Knowledge Base](/docs/concepts/data/knowledge-base.md): The alphaswarm_kb boundary owns AlphaSwarm's cognitive-memory layer. - [KBLayerComposer](/docs/concepts/data/layer-composition.md): KBLayerComposer.compose_recall composes recall across the four - [IMemoryEngine adapter trinity](/docs/concepts/data/memory-engines.md): IMemoryEngine is the vendor-neutral memory control plane. Cognee's - [pgvector control plane](/docs/concepts/data/pgvector-control-plane.md): pgvector is the default BaseVectorStore adapter for the KB layer. - [Hierarchical RAG](/docs/concepts/data/rag.md): HierarchicalRAG lives at alphaswarm_kb.rag.HierarchicalRAG - [Research-papers RAG](/docs/concepts/data/research-papers-rag.md): The research_papers corpus is one of the bundled #### identity AlphaSwarm wraps every identity / token operation in a pluggable - [Federated identity layer](/docs/concepts/identity.md): AlphaSwarm wraps every identity / token operation in a pluggable - [Account integrations](/docs/concepts/identity/account-integrations.md): Per-org credential links the admin operator wires through the - [Account management](/docs/concepts/identity/account-management.md): alphaswarm_admin (internal) note — the internal admin BFF at - [alphaswarm_admin — Microsoft Entra Agent Identity](/docs/concepts/identity/admin-agent-identity.md): Last refreshed: 2026-05-27. - [Auth0 Actions for the AlphaSwarm multi-tenant rollout](/docs/concepts/identity/auth0-actions.md): The Phase 4 enforcement sweep relies on Auth0 to inject - [Auth0 + Microsoft Entra federation runbook](/docs/concepts/identity/auth0-microsoft-federation.md): This runbook covers the one-time operator setup for federating Microsoft Entra ID through Auth0 Universal Login, so AlphaSwarm keeps one identity control plane while still supporting enterprise SSO and account lifecycle features. - [Auth0 setup — comprehensive operator runbook](/docs/concepts/identity/auth0-setup.md): This is the canonical setup guide for AGENTS hard rules 52-55 (the - [Biscuit capability tokens](/docs/concepts/identity/biscuit-capabilities.md): Phase 5 §8.2 — biscuit capability tokens for capability-attenuated agent → MCP delegation. - [Cloud credentials](/docs/concepts/identity/cloud-credentials.md): How AlphaSwarm routes secret resolution through CredentialResolver once the - [Credentials resolver](/docs/concepts/identity/credentials.md): AlphaSwarm collapses every "where does this service's credential come from?" - [Edge authentication & cell routing](/docs/concepts/identity/edge-authentication.md): Every request entering the hosted platform crosses one authentication - [Entra ID as the AlphaSwarm staff user pool](/docs/concepts/identity/entra-internal-tenant.md): Microsoft Entra ID is the first user pool for the managed AlphaSwarm - [AlphaSwarm Management Engine](/docs/concepts/identity/management-engine.md): Canonical narrative for the unified management/control surface - [Microsoft Entra ID (MSAL) setup](/docs/concepts/identity/msal-entra-setup.md): Step-by-step walkthrough for wiring AlphaSwarm's MsalEntraProvider to a - [Multi-tenancy](/docs/concepts/identity/multi-tenancy.md): How AlphaSwarm turns a Microsoft Entra ID tid claim into an - [SCIM Provisioning](/docs/concepts/identity/scim-provisioning.md): AlphaSwarm exposes a SCIM 2.0 provisioning surface at /scim/v2/* for Auth0 - [SPIFFE workload identity](/docs/concepts/identity/spiffe-workload-identity.md): Phase 4 §7.2 — SPIFFE / SPIRE workload identity for AlphaSwarm service-to-service authentication. #### infrastructure - [AlphaSwarm IDE](/docs/concepts/infrastructure/alphaswarm-ide.md): The AlphaSwarm IDE is a white-labeled Eclipse Theia 1.72 distribution + six - [AlphaSwarm IDE roadmap](/docs/concepts/infrastructure/alphaswarm-ide-roadmap.md): This doc maps the [external quant-IDE - [CI/CD pipelines](/docs/concepts/infrastructure/cicd-pipelines.md): The AlphaSwarm AWS deployment is driven by CI/CD: GitHub Actions - [Control-plane topology](/docs/concepts/infrastructure/control-plane-topology.md): Phase 0 of the AlphaSwarm infra-expansion plan. The single source of truth - [IaC runbook](/docs/concepts/infrastructure/iac-runbook.md): "I want to provision X" recipes for the Terraform IaC control plane. - [Kubernetes adapter](/docs/concepts/infrastructure/kubernetes-adapter.md): AlphaSwarm wraps every cluster-side operation in a pluggable - [rpi Kubernetes Deployment](/docs/concepts/infrastructure/kubernetes-rpi-deployment.md): AlphaSwarm deploys to the rpi_kubernetes cluster through the sanctioned - [Service-level view](/docs/concepts/infrastructure/services.md): This page catalogues every service AlphaSwarm runs — the application - [alphaswarm-admin](/docs/concepts/infrastructure/services/alphaswarm-admin.md): Internal-only admin dashboard for AlphaSwarm staff. Audit-first: every - [alphaswarm-beat](/docs/concepts/infrastructure/services/alphaswarm-beat.md): Celery beat process responsible for time-based task dispatch. It writes - [alphaswarm-client](/docs/concepts/infrastructure/services/alphaswarm-client.md): The frontend for local power users — operators running AlphaSwarm on a - [alphaswarm-core](/docs/concepts/infrastructure/services/alphaswarm-core.md): The FastAPI gateway for the AlphaSwarm runtime. Every business route - [alphaswarm-cp](/docs/concepts/infrastructure/services/alphaswarm-cp.md): The standalone control plane. Owns every workload-lifecycle action, - [alphaswarm-docs](/docs/concepts/infrastructure/services/alphaswarm-docs.md): The canonical AlphaSwarm documentation site. Docusaurus + Diátaxis - [alphaswarm-docs-archive](/docs/concepts/infrastructure/services/alphaswarm-docs-archive.md): Sunset documentation archive. Stripe-style: every public-API epoch - [alphaswarm-docs-status](/docs/concepts/infrastructure/services/alphaswarm-docs-status.md): The public status page. Provisioned on Instatus - [alphaswarm-executor](/docs/concepts/infrastructure/services/alphaswarm-executor.md): Celery heavy-compute executor pod — the compute-heavy counterpart of - [alphaswarm-ide](/docs/concepts/infrastructure/services/alphaswarm-ide.md): Browser-tier IDE for AlphaSwarm. White-labeled Theia 1.72 with six - [alphaswarm-ml-mcp](/docs/concepts/infrastructure/services/alphaswarm-ml-mcp.md): Dedicated MCP server publishing the data.ml.* tool slice — Predictor - [alphaswarm-ui](/docs/concepts/infrastructure/services/alphaswarm-ui.md): The cloud-hosted, customer-facing operator dashboard. Auth-gated and - [alphaswarm-website](/docs/concepts/infrastructure/services/alphaswarm-website.md): The public marketing site. Lives in its own repo - [alphaswarm-worker](/docs/concepts/infrastructure/services/alphaswarm-worker.md): Celery orchestration worker pod that drains the light / coordination - [chromadb](/docs/concepts/infrastructure/services/chromadb.md): A vector store used for embedding indices in dev cells and small- - [grafana](/docs/concepts/infrastructure/services/grafana.md): The platform's primary dashboard surface. Bundled with the - [jaeger](/docs/concepts/infrastructure/services/jaeger.md): Distributed tracing backend for the infrastructure trace pipeline — - [loki](/docs/concepts/infrastructure/services/loki.md): Grafana Loki — the log aggregation backend. Receives logs from the - [mlflow](/docs/concepts/infrastructure/services/mlflow.md): The platform's model registry + experiment tracker. Owned by - [neo4j](/docs/concepts/infrastructure/services/neo4j.md): The canonical graph store. Holds the ownership graph (Workstream F), - [otel-collector](/docs/concepts/infrastructure/services/otel-collector.md): The single OTLP ingress for the cluster. Every workload pod sends - [postgres](/docs/concepts/infrastructure/services/postgres.md): The platform's primary OLTP database. Holds every relational table the - [prometheus](/docs/concepts/infrastructure/services/prometheus.md): The cluster-internal metrics scraper. Deployed via - [redis](/docs/concepts/infrastructure/services/redis.md): Multi-purpose key-value store. Holds the kill-switch flag, the BFF - [vector](/docs/concepts/infrastructure/services/vector.md): Vector — the canonical log shipper. Runs as a - [victoriametrics](/docs/concepts/infrastructure/services/victoriametrics.md): VictoriaMetrics — Prometheus-compatible - [Terraform IaC control plane](/docs/concepts/infrastructure/terraform-control-plane.md): Phase 7 of the multi-tenant rollout introduces the 5th sibling - [Worker vs executor images](/docs/concepts/infrastructure/worker-executor-images.md): The AlphaSwarm Celery surface is split into two purpose-built, #### platform - [AlphaSwarm Monorepo Paths](/docs/concepts/platform/alphaswarm-monorepo-paths.md): Status: active. - [Architecture](/docs/concepts/platform/architecture.md): Human entry point. Pair with the AI-agent entry point at - [Class Diagrams](/docs/concepts/platform/class-diagram.md): Pair with alphaswarm_docs/erd.md (database schema) and - [Code Index Governance](/docs/concepts/platform/code-index-governance.md): Status: active. - [Contingency graphs (OCO / OUO / OTO)](/docs/concepts/platform/contingency-graphs.md): Status - [Core Type System](/docs/concepts/platform/core-types.md): Doc map alphaswarmdocs/class-diagram.md#1-symbol--core-enums. - [Domain Model](/docs/concepts/platform/domain-model.md): Doc map alphaswarmdocs/erd.md · Column reference: alphaswarm_docs/data-dictionary.md. - [Entity Graph And Service Control](/docs/concepts/platform/entity-graph-services.md): AlphaSwarm now treats the entity graph as the canonical relationship layer for - [Unified Entity Registry](/docs/concepts/platform/entity-registry.md): The unified entity registry sits on top of the existing - [Entity Relationship Diagram](/docs/concepts/platform/erd.md): Pair with alphaswarm_docs/data-dictionary.md (column-level - [Experiments + Tests umbrella (Phase 1 of the multi-tenant rollout)](/docs/concepts/platform/experiments-tests.md): The umbrella sits above every existing typed run table so the - [Major Flows](/docs/concepts/platform/flows.md): Pair with alphaswarm_docs/architecture.md (system view) and - [Instrument taxonomy](/docs/concepts/platform/instrument-taxonomy.md): Status: Phase 1 shipped (Alembic 0039). Adds REIT / mutual fund / OTC - [Legacy alphaswarm.core.types shim](/docs/concepts/platform/legacy-types-shim.md): Status: Phase 5 finalization shipped. The module is now a thin - [Local platform overlay](/docs/concepts/platform/local-platform.md): Audience: a developer who wants to run AlphaSwarm standalone, without - [Ownership graph (Phase 2 of the multi-tenant rollout)](/docs/concepts/platform/ownership-graph.md): The ownership graph is the projection layer that lets the MCP - [Repository Split](/docs/concepts/platform/repository-split.md): Status: migration guidance. - [AlphaSwarm Scope Catalogue](/docs/concepts/platform/scopes.md): Single source of truth for every authorization scope used by the AlphaSwarm - [Temporal identifier resolution](/docs/concepts/platform/temporal-identifiers.md): Status: Phase 1 shipped (Alembic 0039 + 0040). The #### rl - [Hybrid agentic-RL + backtest](/docs/concepts/rl/agentic-rl.md): AlphaSwarm's port of the FinRL-X "deployment-consistent" blueprint plus the - [RL component reference](/docs/concepts/rl/rl-components.md): This page is a hand-written shortcut. The authoritative source is the - [RL FinAgent Layered Reflection Adapter (Phase 10)](/docs/concepts/rl/rl-finagent.md): Reference docs for the FinAgent multimodal LLM-hybrid agent ported - [Reinforcement learning framework](/docs/concepts/rl/rl-framework.md): The RL layer in AlphaSwarm follows a metaclass-driven, registry-first design - [RL Iceberg data plane](/docs/concepts/rl/rl-iceberg.md): Per-step RL records persist to four Iceberg tables in the namespace - [RL Lab — interactive RL builder](/docs/concepts/rl/rl-lab.md): Lives at /rl/lab in the AlphaSwarm webui. Combines six surfaces into one - [RL Market Dynamics Modeling (Phase 6)](/docs/concepts/rl/rl-market-dynamics.md): Reference docs for the slice-and-merge regime labeller and its - [RL policy backbones](/docs/concepts/rl/rl-policy-backbones.md): Transformer / RNN / Autoencoder / PatchTST feature trunks for the - [RL PRUDEX-Compass Evaluation (Phase 9)](/docs/concepts/rl/rl-prudex-evaluation.md): Reference docs for the PRUDEX-Compass evaluation framework ported - [Weight-centric portfolio pipeline (f_S -> f_A -> f_T -> f_R)](/docs/concepts/rl/weight-centric-pipeline.md): The FinRL-X four-stage protocol that guarantees identical target #### strategy - [Analysis Agents](/docs/concepts/strategy/analysis-agents.md): Three interpretation agents + one deferred reflector. Together they - [Analysis Flows Reference](/docs/concepts/strategy/analysis-flows.md): Framework alphaswarmdocs/analysis-lab.md. - [Analysis Framework](/docs/concepts/strategy/analysis-framework.md): Doc map alphaswarmdocs/analysis-lab.md · Flow reference: alphaswarm_docs/analysis-flows.md. - [Analysis Lab — interactive analysis builder](/docs/concepts/strategy/analysis-lab.md): Backend alphaswarmdocs/analysis-flows.md. - [Backtest engines](/docs/concepts/strategy/backtest-engines.md): Doc map: intro · - [Cross-market arbitrage](/docs/concepts/strategy/cross-market-arbitrage.md): Status: Phase 5 shipped. Combined deliverable across Phase 1 - [Execution paths: WebSocket priority + queue-preserving amendment](/docs/concepts/strategy/execution-paths.md): Status - [Factor Research](/docs/concepts/strategy/factor-research.md): Doc map: alphaswarmdocs/index.md · See alphaswarmdocs/strategy-lifecycle.md for the broader strategy lifecycle. - [HFT / LOB backtest engine](/docs/concepts/strategy/hft-backtest.md): Audience: quants running tick-replay backtests for market-making - [Microstructure toxicity + regime-aware adapter](/docs/concepts/strategy/microstructure-toxicity.md): Audience: anyone running paper or live HFT strategies who wants - [AlphaBacktestExperiment](/docs/concepts/strategy/ml-alpha-backtest.md): The keystone "model used as alpha" experiment — train a model, register - [Graphical ML experiment builder](/docs/concepts/strategy/ml-builder.md): The /ml/builder page composes datasets, preprocessing, model - [Lightweight workbench flows](/docs/concepts/strategy/ml-flows.md): Small synchronous helpers in alphaswarm.ml.flows that - [alphaswarm.ml — native qlib-style ML framework](/docs/concepts/strategy/ml-framework.md): Doc map: alphaswarmdocs/index.md · See alphaswarmdocs/factor-research.md for the alphalens-style evaluation pipeline. - [ML library reference](/docs/concepts/strategy/ml-libraries.md): Per-framework reference for every model wrapper under - [ML preprocessing as data-pipeline nodes](/docs/concepts/strategy/ml-preprocessing-pipeline.md): Bridges alphaswarm.ml.processors into the data - [Interactive ML testing workbench](/docs/concepts/strategy/ml-testing.md): Superseded by strategy-development.md. - [MLOps service inside alphaswarm_models/](/docs/concepts/strategy/mlops-service.md): This page documents the initial MLOps service shipped as additive - [Optimal-control / HJB math layer](/docs/concepts/strategy/optimal-control.md): Audience: quants extending AlphaSwarm with optimal-execution or - [Portfolio options market making — Lucic-Tse 2024-2026](/docs/concepts/strategy/portfolio-options-mm.md): Audience: options market-makers, quant developers writing - [PredictorHub](/docs/concepts/strategy/predictor-hub.md): Status - [Statistical arbitrage primitives](/docs/concepts/strategy/statistical-arbitrage.md): Status - [Strategy Browser](/docs/concepts/strategy/strategy-browser.md): Doc map alphaswarmdocs/strategy-lifecycle.md. - [Strategy Development (Consolidated /strategy-development/*)](/docs/concepts/strategy/strategy-development.md): The Vite frontend exposes a single consolidated umbrella for every - [Strategy Lifecycle](/docs/concepts/strategy/strategy-lifecycle.md): Doc map alphaswarmdocs/flows.md#2-backtest-dispatch. - [Strategy template catalog (Phase 7 of the multi-tenant rollout)](/docs/concepts/strategy/strategy-templates.md): Read-only strategy templates — QuantConnect LEAN's - [vectorbt-pro deep integration](/docs/concepts/strategy/vbtpro-integration.md): Doc map alphaswarmdocs/backtest-engines.md. #### trading - [Observability](/docs/concepts/trading/observability.md): Doc map alphaswarmdocs/flows.md#cross-cutting-progress-bus. - [Observability stack](/docs/concepts/trading/observability-stack.md): Phase 2c + 2d of the AlphaSwarm infra-expansion plan stand up the AlphaSwarm-owned - [Paper Metadata Gate (Strict-Only)](/docs/concepts/trading/paper-metadata-gate.md): Breaking change - [Paper & live trading](/docs/concepts/trading/paper-trading.md): Doc map alphaswarmdocs/flows.md#4-paper-trading-session. - [webui — Next.js 15 frontend](/docs/concepts/trading/webui.md): Doc map alphaswarmdocs/architecture.md#system-component-diagram. ### how-to #### alphaswarm-admin-entra-setup End-to-end procedure for connecting the alphaswarm_admin service (backend - [Wire alphaswarm_admin against the AlphaSwarm staff Entra tenant](/docs/how-to/alphaswarm-admin-entra-setup.md): End-to-end procedure for connecting the alphaswarm_admin service (backend #### audit-lake-reconstruction Phase 7 §10 (RESTRUCTURING_PLAN.md) — operating procedure for the - [Audit lake reconstruction runbook](/docs/how-to/audit-lake-reconstruction.md): Phase 7 §10 (RESTRUCTURING_PLAN.md) — operating procedure for the #### cell-data-plane-migration Phase 6 §9 (RESTRUCTURING_PLAN.md) — operating procedure for - [Cell data plane migration runbook](/docs/how-to/cell-data-plane-migration.md): Phase 6 §9 (RESTRUCTURING_PLAN.md) — operating procedure for #### cell-router-cutover Phase 3 §6.4 runbook — deploy the Envoy cell router + alphaswarm-tenant-router decision service, canary 10% / 50% / 100%, retire the Python FastAPI proxy. - [Cell-router cutover runbook](/docs/how-to/cell-router-cutover.md): Phase 3 §6.4 runbook — deploy the Envoy cell router + alphaswarm-tenant-router decision service, canary 10% / 50% / 100%, retire the Python FastAPI proxy. #### chainguard-base-migration Phase 2 §5.1 runbook — migrate AlphaSwarm Dockerfiles to cgr.dev/chainguard/* bases, verify cosign keyless signatures + SBOMs, ratchet Kyverno from Audit to Enforce. - [Chainguard base migration runbook](/docs/how-to/chainguard-base-migration.md): Phase 2 §5.1 runbook — migrate AlphaSwarm Dockerfiles to cgr.dev/chainguard/* bases, verify cosign keyless signatures + SBOMs, ratchet Kyverno from Audit to Enforce. #### entra-onboard-new-staff Procedure for adding a new AlphaSwarm employee to the company's Entra - [Onboard a new staff member into Entra](/docs/how-to/entra-onboard-new-staff.md): Procedure for adding a new AlphaSwarm employee to the company's Entra #### entra-rotate-secrets The AlphaSwarm staff Entra rollout aims for zero stored secrets — CI - [Rotate Entra ID app secrets](/docs/how-to/entra-rotate-secrets.md): The AlphaSwarm staff Entra rollout aims for zero stored secrets — CI #### entra-terraform-bootstrap Step-by-step procedure for taking the AlphaSwarm staff Microsoft Entra ID - [Bootstrap the AlphaSwarm Entra ID staff tenant](/docs/how-to/entra-terraform-bootstrap.md): Step-by-step procedure for taking the AlphaSwarm staff Microsoft Entra ID #### linkerd-spire-rollout Phase 4 §7.1 / §7.2 runbook — install per-cell Linkerd 2.16 + SPIRE 1.10, validate mTLS, ratchet from Audit to Enforce. - [Linkerd + SPIRE rollout runbook](/docs/how-to/linkerd-spire-rollout.md): Phase 4 §7.1 / §7.2 runbook — install per-cell Linkerd 2.16 + SPIRE 1.10, validate mTLS, ratchet from Audit to Enforce. #### mlops - [Cross-repo lineage bridge](/docs/how-to/mlops/cross-repo-lineage.md): The agentic_assistants repository maintains a shared lineage graph - [Kubernetes deployment](/docs/how-to/mlops/k8s-deployment.md): AlphaSwarm ships Kustomize manifests under alphaswarmplatform/deploy/k8s/base/ - [Model serving](/docs/how-to/mlops/serving.md): AlphaSwarm ships three serving adapters. All three share the same #### operations - [Adding a new InfrastructureProvider to alphaswarm_controller](/docs/how-to/operations/add-new-provider.md): Step-by-step guide for shipping a new InfrastructureProvider implementation (AGENTS rule 45 / ADR 004). - [AlphaSwarm.FUND Blue/Green Cutover](/docs/how-to/operations/alphaswarm-fund-blue-green-cutover.md): Runbook for migrating alphaswarm.fund traffic to the tower cluster with a short, - [Auth0 checklist for Kubernetes login](/docs/how-to/operations/auth0-k8s-checklist.md): This checklist captures the Auth0-side changes required for the Kubernetes - [AWS Hybrid Deployment Guide](/docs/how-to/operations/aws-deploy.md): Companion runbook: aws-runbook.md. - [AWS Minimum Tier Rollback Playbook](/docs/how-to/operations/aws-minimum-rollback.md): Companion to - [Single-Account Minimum AWS Deployment](/docs/how-to/operations/aws-minimum-single-account.md): Companion docs: aws-deploy.md for the full - [AWS Hybrid Operational Runbook](/docs/how-to/operations/aws-runbook.md): Companion to aws-deploy.md. Page this when the - [Bot Canary Rollout Playbook](/docs/how-to/operations/bot-canary-rollout-playbook.md): When to use it, how to read the dashboards, how to abort, and how - [Operations runbook — CI/CD deploy](/docs/how-to/operations/cicd-deploy.md): Task-oriented steps for the AlphaSwarm AWS CI/CD pipeline. For the - [Cloud-CLI temporary credentials](/docs/how-to/operations/cloud-cli-temporary-credentials.md): How to use the CloudCliCredentialWizard in the admin Settings - [Operations runbook — Configuration management](/docs/how-to/operations/configuration-management.md): How env vars, ConfigMaps, and Secrets flow through the AlphaSwarm stack. - [Connect a company cloud account](/docs/how-to/operations/connect-company-cloud-account.md): The cloud onboarding wizard in alphaswarm_admin (route - [Control the platform ECS deployment](/docs/how-to/operations/control-platform-ecs-deployment.md): The Platform page in alphaswarm_admin (route /platform, API - [Operations runbook — Edge deployment](/docs/how-to/operations/edge-deploy.md): Deploying AlphaSwarm to edge / on-prem locations where the standard cloud K8s overlays don't fit. - [Go-live: minimum deployment](/docs/how-to/operations/go-live-minimum.md): Deep-dive companions: aws-deploy.md (full hybrid - [HFT Node Onboarding](/docs/how-to/operations/hft-node-onboarding.md): How to bring up a new dedicated node for Frequency.HFT bots. - [Operations runbook — Incident response](/docs/how-to/operations/incident-response.md): Standard playbook for diagnosing + recovering from AlphaSwarm production incidents. - [Kill Switch Incident Response](/docs/how-to/operations/kill-switch-incident-response.md): Three-scope kill switch (bot / fleet / platform). Quarterly drill - [Operations runbook — Kubernetes deployment](/docs/how-to/operations/kubernetes-deploy.md): End-to-end walkthrough for shipping AlphaSwarm to any Kubernetes cluster (EKS, - [Operations runbook — Local setup](/docs/how-to/operations/local-setup.md): This walks a brand-new developer from git clone to a running local AlphaSwarm stack. - [Operations runbook — Secret rotation](/docs/how-to/operations/rotate-secrets.md): Zero-downtime credential rotation for the AlphaSwarm control plane + workloads. - [RTS 6 / SEC 15c3-5 Annual Validation Report](/docs/how-to/operations/rts6-validation-report-generation.md): Mechanical generation + sign-off workflow. - [Tower Two-Node Cluster Deploy](/docs/how-to/operations/tower-cluster-deploy.md): Deploy AlphaSwarm to the dedicated two-node cluster (alphaswarm-tower control plane + #### per-tenant-mcp-rollout Phase 5 §8.1 runbook — deploy per-tenant MCP servers + the gVisor agent-sandbox pool + Cell-Bound-Authorization validation. - [Per-tenant MCP rollout runbook](/docs/how-to/per-tenant-mcp-rollout.md): Phase 5 §8.1 runbook — deploy per-tenant MCP servers + the gVisor agent-sandbox pool + Cell-Bound-Authorization validation. #### recipes Task-oriented, results-first. Each recipe answers a single - [Recipes](/docs/how-to/recipes.md): Task-oriented, results-first. Each recipe answers a single - [Recipe: add a strategy](/docs/how-to/recipes/add-a-strategy.md): The 5-minute happy path: - [Recipe: promote a bot to paper](/docs/how-to/recipes/promote-a-bot-to-paper.md): The paper-metadata gate - [Recipe: query data via MCP](/docs/how-to/recipes/query-data-via-mcp.md): AGENTS rule 22: agents NEVER read Postgres or Iceberg directly. - [Recipe: run a backtest from YAML](/docs/how-to/recipes/run-a-backtest-from-yaml.md): Choose your engine - [Recipe: snapshot an agent spec](/docs/how-to/recipes/snapshot-an-agent-spec.md): The response carries spec_hash and version. If you change a #### runbooks - [Runbook — Disaster Recovery: full restore (under 30 min)](/docs/how-to/runbooks/dr-restore.md): Restores the Phase 6 reliability surface from S3 in three layers. - [Runbook — QuestDB WAL apply stall](/docs/how-to/runbooks/questdb-wal-stall.md): Symptoms: - [Runbook — quota-exhaustion](/docs/how-to/runbooks/quota-exhaustion.md): A bucket has fired AQPRatelimitBucketAt80Percent, - [Runbook — dbt snapshot deadlock](/docs/how-to/runbooks/snapshot-deadlock.md): Symptoms: #### tenant-router-auth-rollout Configure fail-closed JWT verification on the alphaswarm-tenant-router edge — issuer/audience/JWKS wiring, permissive canary, enforcement, CBA keys, verification checks, and rollback. - [Tenant-router auth rollout runbook](/docs/how-to/tenant-router-auth-rollout.md): Configure fail-closed JWT verification on the alphaswarm-tenant-router edge — issuer/audience/JWKS wiring, permissive canary, enforcement, CBA keys, verification checks, and rollback. ### intro Triple-axis table of contents for the AlphaSwarm docs. - [Documentation Index](/docs/intro.md): Triple-axis table of contents for the AlphaSwarm docs. #### conventions Frontmatter is mandatory - [Conventions](/docs/intro/conventions.md): Frontmatter is mandatory #### glossary Project-specific jargon used across AlphaSwarm, with a definition and a pointer - [Glossary](/docs/intro/glossary.md): Project-specific jargon used across AlphaSwarm, with a definition and a pointer #### installation This page documents the install-time requirements for AlphaSwarm and its - [Installation](/docs/intro/installation.md): This page documents the install-time requirements for AlphaSwarm and its #### quickstart Target: a fresh checkout of alphaswarm to a green backtest - [Quickstart](/docs/intro/quickstart.md): Target: a fresh checkout of alphaswarm to a green backtest #### repository-orientation AlphaSwarm is a monorepo organised by responsibility. The boundary between - [Repository orientation](/docs/intro/repository-orientation.md): AlphaSwarm is a monorepo organised by responsibility. The boundary between ### operations #### break-glass Procedure for assuming the AqpAdminBreakGlassRole during an - [Break-glass runbook](/docs/operations/break-glass.md): Procedure for assuming the AqpAdminBreakGlassRole during an #### dr-replay Disaster-recovery rehearsal procedure for AlphaSwarm. Targets: - [DR replay runbook](/docs/operations/dr-replay.md): Disaster-recovery rehearsal procedure for AlphaSwarm. Targets: #### multi-account-rollout The Phase 4 Control Tower + cross-account IAM + dev->staging - [Multi-account rollout runbook](/docs/operations/multi-account-rollout.md): The Phase 4 Control Tower + cross-account IAM + dev->staging ### reference #### api This page is auto-generated from - [API reference](/docs/reference/api.md): This page is auto-generated from - [Liveness probe.](/docs/reference/api/health/get-livez.md): Liveness probe. - [Readiness probe — confirms migrations applied + downstreams reachable.](/docs/reference/api/health/get-readyz.md): Readiness probe — confirms migrations applied + downstreams reachable. #### data-dictionary Pair with alphaswarm_docs/erd.md (visual schema) and - [Data Dictionary](/docs/reference/data-dictionary.md): Pair with alphaswarm_docs/erd.md (visual schema) and #### manage-api This is the alphaswarm_controller surface at manage.alpha-swarm.ai. It is - [Control-plane API](/docs/reference/manage-api.md): This is the alphaswarm_controller surface at manage.alpha-swarm.ai. It is - [Control-plane liveness probe.](/docs/reference/manage-api/workloads/get-manage-livez.md): Control-plane liveness probe. #### python This tree is auto-generated by - [Python reference](/docs/reference/python.md): This tree is auto-generated by - [alphaswarm](/docs/reference/python/alphaswarm.md): This page would normally be auto-generated by mdxify from the - [alphaswarm_bots](/docs/reference/python/alphaswarm_bots.md): This page would normally be auto-generated by mdxify from the - [alphaswarm_controller](/docs/reference/python/alphaswarm_controller.md): This page would normally be auto-generated by mdxify from the - [alphaswarm_core](/docs/reference/python/alphaswarm_core.md): This page would normally be auto-generated by mdxify from the - [alphaswarm_models](/docs/reference/python/alphaswarm_models.md): This page would normally be auto-generated by mdxify from the - [alphaswarm_rl](/docs/reference/python/alphaswarm_rl.md): This page would normally be auto-generated by mdxify from the ### release-notes Customer-facing release notes for the AlphaSwarm. New - [Release notes](/docs/release-notes.md): Customer-facing release notes for the AlphaSwarm. New #### 2026-06-01-initial-release This release marks two significant changes for AlphaSwarm customers: - [Release 2026-06-01 — Docs migration + first API epoch](/docs/release-notes/2026-06-01-initial-release.md): This release marks two significant changes for AlphaSwarm customers: ### tutorials Runnable, learning-oriented walkthroughs. Each tutorial assumes the - [Tutorials](/docs/tutorials.md): Runnable, learning-oriented walkthroughs. Each tutorial assumes the #### first-agent-workflow Goal: stand up a three-node agentic loop driven by WorkflowRuntime, - [Your first agent workflow](/docs/tutorials/first-agent-workflow.md): Goal: stand up a three-node agentic loop driven by WorkflowRuntime, #### first-backtest Goal: from blank slate to a backtest with a non-zero Sharpe on - [Your first backtest](/docs/tutorials/first-backtest.md): Goal: from blank slate to a backtest with a non-zero Sharpe on #### first-bot Goal: take the strategy from first-backtest - [Your first bot](/docs/tutorials/first-bot.md): Goal: take the strategy from first-backtest #### first-paper-trading-session Goal: drive a paper-trading session from the bot you authored in - [Your first paper trading session](/docs/tutorials/first-paper-trading-session.md): Goal: drive a paper-trading session from the bot you authored in #### first-rl-experiment Goal: from blank RLExperimentSpec to a trained PPO agent with - [Your first RL experiment](/docs/tutorials/first-rl-experiment.md): Goal: from blank RLExperimentSpec to a trained PPO agent with ## search - [Search the documentation](/search.md)