Back to corpus
architecturetechnical paper candidatescore 18

Stage 4: ARCHITECTURE — Fleet Evolution Engine v2

``` DISCOVER (KARL-ranked cohort selection) ↓ DISPATCH (Python orchestrator → AuraGateway skill command) ↓ PARSE (structured output → Supabase tasks) ↓ REVIEW (Codex adversarial → issue blocks → Supabase tasks) ↓ EVOLVE (Hydra cycles → Swift bridge polls quality gate) ↓ SHIP (quality gate passes → auto TestFlight trigger) ↓ LEARN (run as KARL trajectory → improve next cohort routing) ```

Open full HTMLDownload HTMLCorpus JSONPaper schema

Full HTML reader

Read the full artifact

Open in new tab

Extracted abstract or opening context

# Stage 4: ARCHITECTURE — Fleet Evolution Engine v2 > Forged from: Track 2 (Python orchestrator spine) + Track 3 (Hydra bridge) + Entanglement seeds **Responsibility:** Own the pipeline. Advance stages. Parse output. Handle retries. **Prefect scheduling:** Every 5 minutes, check queue and start up to `maxConcurrent` apps (read from Supabase config). **Tier 1 (structured block):** Scan output for `<!-- EVOLUTION_TASKS_START -->` JSON block. If found, use it directly. **Tier 2 (LLM extraction):** If no structured block, pass output to Gemini Flash with extraction prompt:

Promotion decision

What has to happen next

Promote into a technical note or architecture paper with implementation anchors.

Why this is not always a full paper yet

Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.