docs: record protected gpu model discovery
CI / test (push) Successful in 1m7s
CI / test (pull_request) Successful in 1m5s

This commit is contained in:
m
2026-08-23 11:00:00 +03:00
parent b4d166480e
commit 97ed2e90ee
4 changed files with 55 additions and 3 deletions
+1 -1
View File
@@ -5,7 +5,7 @@
- `CURRENT_PRODUCTION_STATUS = LIVE_HEALTHY`; public `/healthz` is healthy. The exact deployed revision is authoritative in the deployment log and merge record, not duplicated in this mutable status projection. Evidence: `docs/evidence/audit-runtime-status-2026-08-23-v1.json`.
- `FAST_PROVIDER_STATUS = LIVE_SMOKE_SUPPORTED`: local CPU gateway authentication is configured through a protected file reference; four bounded local-only advisory cases completed through cognitive-worker. Evidence: `docs/evidence/wp-601-fast-gateway-live-smoke-2026-08-22-v1.json`.
- `MEDIUM_PROVIDER_STATUS = LIVE_BOUNDED_SMOKE_SUPPORTED_UNPROMOTED`; four bounded local-provider cases passed with template thinking disabled. This is availability/response-shape evidence only; lifecycle qualification, activation and promotion remain unclaimed. Evidence: `docs/evidence/medium-live-bounded-smoke-2026-08-23-v1.json`.
- `HEAVY_PROVIDER_STATUS = LIVE_QUALIFIED_UNACTIVATED`; the qualified GPU binding remains inactive. Evidence: `docs/evidence/wp-200-heavy-qualification-2026-08-22-v2.json`.
- `HEAVY_PROVIDER_STATUS = LIVE_DISCOVERED_AVAILABLE_UNPROMOTED`; protected GPU discovery currently sees two available models, both `DISCOVERED`/`INACTIVE` with capability and invokability not yet observed. GPU availability is not treated as intelligence or routing priority. Evidence: `docs/evidence/gpu-discovery-2026-08-23-v1.json`.
- `AGENT_EXECUTION_STATUS = DURABLE_DRY_RUN_CANARY_VERIFIED`; the explicit production canary persisted a dispatch intent, an ACTIVE lease and callback head sequence 2. No independently operated bridge endpoint is configured, so bridge-backed execution remains unavailable. Evidence: `docs/evidence/agent-execution-dry-run-canary-2026-08-23-v1.json`.
- `FLEET_EXECUTION_EVIDENCE = docs/evidence/audit-runtime-status-2026-08-23-v1.json`; FAST, MEDIUM and LIFEBOAT are ready, HEAVY is explicitly degraded by provider timeout, FAST remains preferred/selected, and LIFEBOAT is the fallback tier.
- `TRUTH_RESEARCH_STATUS = LIVE_CANARY_PERSISTED`; the web research canary preserved an OBSERVED/COMPLETE/NO-CONFLICT source chain while leaving the claim UNVERIFIED and belief PROVISIONAL. Evidence: `docs/evidence/truth-runtime-canary-2026-08-23-v1.json`.
@@ -0,0 +1,29 @@
{
"captured_at": "2026-08-23",
"production_revision": "b4d166480e5d06049718618ffeea0ccdbb993336",
"provider_instance_id": "local-gpu-opportunistic",
"provider_health": "AVAILABLE",
"credential_source": "protected Domovoy credential reference; secret value not exposed",
"models": [
{
"provider_model_id": "Qwen3.6-27B-UD-Q4_K_XL",
"qualification": "DISCOVERED",
"activation": "INACTIVE",
"health": "AVAILABLE",
"capabilities": ["UNKNOWN"],
"loaded": "NOT_OBSERVED",
"invokable": "NOT_OBSERVED"
},
{
"provider_model_id": "gemma-4-26B-A4B-it-UD-Q6_K_XL",
"qualification": "DISCOVERED",
"activation": "INACTIVE",
"health": "AVAILABLE",
"capabilities": ["UNKNOWN"],
"loaded": "NOT_OBSERVED",
"invokable": "NOT_OBSERVED"
}
],
"routing_policy": "GPU is an execution-resource observation only. Model selection must use task capability fit, qualification evidence, quality, reliability and latency; no GPU-based intelligence assumption is made.",
"secrets_exposed": false
}