SCBE Full System Layer Map (Kernel -> AI -> Swarm -> Cloud -> Training)
Last updated: 2026-02-17
Scope
This map consolidates the current SCBE ecosystem across core repositories under issdandavis and documents what is implemented, partial, or conceptual.
Layer Stack (authoritative operational view)
Layer 0: Canonical Kernel Contract
- Purpose: protocol authority and term discipline.
- Status: implemented.
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
SPEC.mdCONCEPTS.md
Layer 1: SCBE Governance Runtime (14-layer core)
- Purpose: deterministic ALLOW/QUARANTINE/DENY decisions.
- Status: implemented.
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
api/main.pyREADME.mdARCHITECTURE.mddocs/hydra/ARCHITECTURE.md(SCBE as Layer 1 in HYDRA stack)
Layer 2: GeoSeal / Mixed-Curvature Access Kernel
- Purpose: geometric trust fusion and quarantine/memory-write gating.
- Status: implemented (v2 primitives present).
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
src/geoseal.pysrc/geoseal_v2.pytests/test_geoseal.pytests/test_geoseal_v2.py
Layer 3: Single-Agent Browser Execution Plane (AetherBrowse)
- Purpose: governed browser execution with containment checks.
- Status: implemented (branch-integrated operational tooling).
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
agents/browser/main.pyagents/aetherbrowse_cli.pydocs/AETHERBROWSE_GOVERNANCE.md
Layer 4: Multi-Agent Browser/Task Execution (Swarm Runner)
- Purpose: run 2-5+ governed jobs with verification and per-job decision records.
- Status: implemented.
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
scripts/aetherbrowse_swarm_runner.pyexamples/aetherbrowse_tasks.sample.jsonschemas/decision_record.schema.json
Layer 5: HYDRA Coordination + Governed Swarm
- Purpose: head/limb/librarian/ledger orchestration above SCBE kernel.
- Status: partial (reference architecture documented; runtime pieces distributed).
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
docs/hydra/ARCHITECTURE.mdagents/swarm_browser.pytraining/doc_manifest.json(roundtable-style verification metadata)
Layer 6: Workflow Orchestration Connectors (Asana + n8n)
- Purpose: scheduled external tasks -> governed browser actions -> feedback loop.
- Status: implemented.
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
scripts/asana_aetherbrowse_orchestrator.pyscripts/n8n_aetherbrowse_bridge.pyworkflows/n8n/asana_aetherbrowse_scheduler.workflow.jsondocs/ASANA_AETHERBROWSE_AUTOMATION.mddocs/N8N_AETHERBROWSE_INTEGRATION.md
Layer 7: Cloud Execution Plane
- Purpose: remote headless execution for low-spec local machines.
- Status: implemented.
- Primary repo:
SCBE-AETHERMOORE. - Source surfaces:
deploy/gcloud/deploy_aetherbrowse.shdeploy/gcloud/Dockerfile.aetherbrowsedocs/AETHERBROWSE_CLOUD_RUN.md
Layer 8: Data/Training Plane (Notion -> JSONL -> HF -> Vertex)
- Purpose: convert system knowledge and operation traces into trainable datasets/models.
- Status: implemented (with secrets/config dependency).
- Primary repos:
SCBE-AETHERMOORE,phdm-21d-embedding. - Source surfaces:
scripts/notion_access_check.pyscripts/notion_to_dataset.pyscripts/push_to_hf.py.github/workflows/notion-to-dataset.yml.github/workflows/huggingface-sync.yml.github/workflows/vertex-training.ymlphdm-21d-embedding/scripts/markdown_to_jsonl.pyphdm-21d-embedding/scripts/push_jsonl_dataset.py
Layer 9: Ecosystem/Protocol Extensions (Spiralverse + Security Gate)
- Purpose: AI-to-AI protocol + hardened gate modules.
- Status: implemented as satellite repos.
- Primary repos:
spiralverse-protocol,scbe-security-gate. - Source surfaces:
spiralverse-protocol/README.mdscbe-security-gate/README.md
Layer 10: Space Station / Agentic Swarm Factory Narrative Plane
- Purpose: mission-control metaphor and orbital governance framing.
- Status: conceptual + documentation heavy (not yet full runtime layer).
- Primary repo:
scbe-aethermoore-demo. - Source surfaces:
scbe-aethermoore-demo/L1_AETHERMOORE_STATION.md
Core Repository Roles
| Repo | Role in stack | Current role |
|---|---|---|
SCBE-AETHERMOORE | canonical kernel + ops plane | primary production repo |
phdm-21d-embedding | training data conversion + HF push | data pipeline utility |
aws-lambda-simple-web-app | mirrored SCBE docs/spec package | secondary source; consolidate selectively |
Entropicdefenseengineproposal | proposal/prototype surfaces | extract unique docs and merge to canonical |
spiralverse-protocol | AI-to-AI protocol extension | keep as module repo |
scbe-security-gate | standalone gate hardening | keep as module repo |
scbe-aethermoore-demo | narrative/demo shell | keep for demos; avoid canonical drift |
Readiness by objective
| Objective | Status | Notes |
|---|---|---|
| Governed single-agent execution | ready | AetherBrowse + containment checks available |
| Governed multi-job automation | ready | swarm runner + verification + decision records |
| Asana scheduled automation | ready | orchestration script + n8n workflow available |
| Cloud browser operations | ready | Cloud Run deploy artifacts present |
| Notion -> HF dataset flow | ready with config | requires NOTION_API_KEY and HF_TOKEN secrets |
| Vertex/GKE model training handoff | ready with config | requires GCP_SA_KEY and project configuration |
| Full space-station swarm factory runtime | partial | conceptual framing exists; needs mission runtime implementation |
Highest-value next consolidation actions
- Declare
SCBE-AETHERMOOREas sole canonical runtime and spec authority. - Import only unique docs/code from satellite repos; do not duplicate whole trees.
- Promote AetherBrowse + Asana orchestration from feature branch into mainline release.
- Add one mission-runtime package (
mission_control/) that operationalizes Layer 10 concepts into runnable workflows.
Manifest contract and validation
- Machine-readable manifest:
docs/scbe_full_system_layer_manifest.json. - Manifest schema:
docs/scbe_full_system_layer_manifest.schema.json. - Validator script:
scripts/validate_layer_manifest.py. - CI workflow gate:
.github/workflows/validate-layer-manifest.yml. - Execution-order runbook:
docs/RUNBOOK.md.
Known drift to control
- Formula notation drift across docs (
H(d,R)variants) needs one canonical expression policy. - Duplicate architecture prose across repos increases training inconsistency.
- Secret-dependent workflows fail silently if GitHub secrets are absent.