Norm

Fleet Intel Pipeline

Marco fleet panel -- Vibe to DynaRAG to Carousel to Holon intent pipeline with job polling.

Fleet Intel Pipeline (v2026-09-30T111455Z-5e473105f4) 2026-09-30 | Marco fleet panel -- Vibe to DynaRAG to Carousel to Holon intent pipeline with job polling. 1 - API ENDPOINTS - Handling API requests POST /api/fleet/intel/intent daemon/src/fleet_intel_router.py IntentRequest IntentResponse GET /api/fleet/intel/status daemon/src/fleet_intel_router.py fleet_status services_task cost_task GET /api/fleet/intel/metrics daemon/src/fleet_intel_router.py fleet_metrics cost_raw holon_throughput intent request status snapshot 2 - PIPELINE PROCESSING - Processing intent requests start_pipeline daemon/src/fleet_intel_router.py vibe_request daemon/src/fleet_intel_router.py _post GET /api/vibe/analyze dynarag_request daemon/src/fleet_intel_router.py _post POST /api/dynarag/generate carousel_request daemon/src/fleet_intel_router.py _post POST /api/carousel/summarize holon_request daemon/src/fleet_intel_router.py _post POST /api/holon/create vibe analysis dynarag genera… carousel summa… holon creation 3 - JOB TRACKING - Managing intent jobs _jobs daemon/src/fleet_intel_router.py _jobs _jobs_lock _MAX_JOBS reset_locks daemon/src/fleet_intel_router.py reset_locks _jobs_lock _pipeline_semaphore _prune_jobs daemon/src/fleet_intel_router.py _prune_jobs _jobs _MAX_JOBS job tracking lock reset 1. Receive intent request 2. Start pipeline 3. Analyze with Vibe 4. Generate with DynaRAG 5. Summarize with Carousel 6. Create Holon 7. Update job status IntentRequest text: 'Analyze the current fleet status' context: 'Fleet Intel Panel' tags: ['fleet', 'status', 'analysis'] create_holon: true IntentResponse ok: true job_id: '123e4567-e89b-12d3-a456-426614174000' vibe_id: 'vibe-123' dynarag_topic: 'dynarag-123' status: 'queued' The pipeline processing steps are sequential and depend on the previous step's response. The job tracking system uses an in-memory dictionary to manage job states. Legend actor = initiates work · process = code path · store = state on disk · bus = durable queue · dashed = separate process cyan = request flow · pink = state read/write · dashed green = pull / return path · red = refusal As of 2026-09-30 11:14 UTC, from commit 5e47310

Provenance

This drawing is generated, not drawn. It is rebuilt from the source files below, so when they change the picture changes — a diagram here cannot quietly describe a system that no longer works this way.

Owner
Norm
Slug
fleet-intel-pipeline
Rendered
2026-09-30T111455Z
From commit
5e473105f4
Updated
30 Sep 2026, 11:14 UTC
Sources at
5e47310
Watches
1 path
  • daemon/src/fleet_intel_router.py