Triage useful ideas from disconnected machines before promotion. Not live truth — a queue you park, archive, restore, or promote. Thinktank bridge: bridge runbook
Thinktank bridge P1: /operator/ideas thinktank lane + promotion concept publish… — Thinktank bridge P1: /operator/ideas thinktank lane + promotion concept published v0.7.3. Tag ACL proposal only — no mutation.
Fable curation pass 2026-07-09 (post-mesh pivot). Scanned 343 AI-visible catalo… — Fable curation pass 2026-07-09 (post-mesh pivot). Scanned 343 AI-visible catalog entries, read 27 context packets, ranked 5 early-stage project picks. Recommended first: ecosystem capability card repair (ghost product-c…
Operator correction (2026-07-09): VPS loopback 127.0.0.1:8081 is uapi-explorer.… — Operator correction (2026-07-09): VPS loopback 127.0.0.1:8081 is uapi-explorer.service (UAI Explorer public catalog v0.7.0), NOT a cloud-GPU broker. Beacon proxies explorer.united-ai.ca to this port. Prior operator rece…
2026-07-05 human docs reorg audit (owner feedback: documentation mess, jumbled … — 2026-07-05 human docs reorg audit (owner feedback: documentation mess, jumbled UX). Top pain points for human navigation: 1) /docs renders flat 407-row wall (~310KB) with no default filter — nav label "Runbooks" mislea…
2026-07-04 wipe session documentation pass COMPLETE. Handoffs index: [REDACTED:… — 2026-07-04 wipe session documentation pass COMPLETE. Handoffs index: [REDACTED:high-entropy].md. Session CLOSED: Stage 2 VPS DONE, Stage 2.5 Proxmox reboot (Echo OK, Director DOWN), Cadence Fable DEPLOYED, START_HERE + …
Governance vNext now includes no-local-copies / local_data_hoarding=never rule … — Governance vNext now includes no-local-copies / local_data_hoarding=never rule (2026-07-04). Consider surfacing in bootstrap required_rules or start-here MOC card for agent visibility.
Trammel preserve published: product + index; preserved_partial; Beacon inactive… — Trammel preserve published: product + index; preserved_partial; Beacon inactive route retained. Gaps: VPS static roots missing pre-rescue.
Observation: operator starting new product Pipfitter — iOS smart spirit level l… — Observation: operator starting new product Pipfitter — iOS smart spirit level leveraging phone sensors, Stabila-inspired UX (audio guidance, REF mode, mm/m and degree readouts). Research phase complete; no catalog entry…
Mac-side clean-slate-2026-07-01 handoff documents are not synced to Explorer. D… — Mac-side clean-slate-2026-07-01 handoff documents are not synced to Explorer. During a 2026-07-02 long-pass session on the VPS (/root/uapi-service-lab), the operator-referenced Mac paths (/Users/scott/Documents/New [RED…
Codex B70 preinstall update: Forge is confirmed as target node 10.0.0.248; Echo… — Codex B70 preinstall update: Forge is confirmed as target node 10.0.0.248; Echo is 10.0.0.249; CT103 Engineering is 10.0.0.103. HomeLab was read-only checked as quorate 3 nodes with HA idle at observation time. Forge cu…
Delivered lightweight OpenAI-compatible supergrok relay service for lab/Thinkta… — Delivered lightweight OpenAI-compatible supergrok relay service for lab/Thinktank use. Alpine container. Supports specialty agents (planner, coder, thinktank, uai). Includes groupchat.py for multi-agent orchestration. F…
Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canon… — Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canonical `uai-thinktank`-tagged context remains forbidden to this AI lane. This is documentation visibility only and grants no runtime, provider…
UAI Thinktank Explorer publication intake packet is ready for review. Product:… — UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canon… — Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canonical `uai-thinktank`-tagged context remains forbidden to this AI lane. This is documentation visibility only and grants no runtime, provider…
UAI Thinktank Explorer visibility check remains blocked after promotion. Promot… — UAI Thinktank Explorer visibility check remains blocked after promotion. Promoted slug uai-doc-uai-thinktank-ai-context returned context status 403 to the current AI lane. The promoted item requires the uai-thinktank ta…
UAI Thinktank Explorer publication intake packet is ready for review. Product:… — UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
UAI GPU Cloud + UAPI Bridge plan ingest reached Explorer feedback and promotion… — UAI GPU Cloud + UAPI Bridge plan ingest reached Explorer feedback and promotion, but the promoted context is not readable by this AI lane. Target slug: [REDACTED:high-entropy]. Promotion request id: 196. Catalog item id…
proved sidecars can paste bounded Codex, GLM, Grok, browser-relay, or local-w… — proved sidecars can paste bounded Codex, GLM, Grok, browser-relay, or local-worker critique outputs into the External Result Intake card, but the proposal must still treat those outputs as review evidence, not imp…
comparison sources without ranking official docs versus anecdotal/community evi… — comparison sources without ranking official docs versus anecdotal/community evidence. That matters for Intel Arc Pro B70 and scheduler maturity. - The embedded self-review appendix is valuable but awkward. It criticizes…
x. Some of those critiques are sharper than the main body, especially on schema… — x. Some of those critiques are sharper than the main body, especially on schema incompleteness, storage pressure, and K8s-on-Proxmox risk. What I do not like: - The opening "Connection TL;DR" is a serious hygiene fail…
remote artifact storage: https://mlflow.[REDACTED:high-entropy]/ - MLflow artif… — remote artifact storage: https://mlflow.[REDACTED:high-entropy]/ - MLflow artifact stores: https://mlflow.[REDACTED:high-entropy]/ - MinIO AIStor container/S3-family docs: https://docs.min.io/aistor/installation/contain…
https://docs.vllm.ai/ - vLLM OpenAI-compatible server: https://docs.vllm.[REDAC… — https://docs.vllm.ai/ - vLLM OpenAI-compatible server: https://docs.vllm.[REDACTED:high-entropy].html - vLLM Kubernetes/Helm deployment docs: https://docs.vllm.ai/en/stable/deployment/frameworks/helm.html - Ollama API d…
Runbook, fetched through AI context slug `[REDACTED:high-entropy]`. - UAI Devel… — Runbook, fetched through AI context slug `[REDACTED:high-entropy]`. - UAI Development Standards Runbook, fetched through AI context slug `[REDACTED:high-entropy]`. - UAI Agent Feedback Convention, fetched through AI con…
ng | LiteLLM belongs in the model-provider/gateway layer, not the hardware sche… — ng | LiteLLM belongs in the model-provider/gateway layer, not the hardware scheduler; budgets should be mirrored in UAPI policy. | | OpenTelemetry | The OpenTelemetry Collector is a vendor-agnostic receiver/processor/ex…
nit of admission for applications that run to completion, and AdmissionChecks a… — nit of admission for applications that run to completion, and AdmissionChecks allow extra criteria before admission. Sources: https://kueue.sigs.k8s.io/docs/concepts/workload/ and https://kueue.sigs.k8s.io/docs/concepts…
down | | GPU sharing used for sensitive workloads | Critical | Medium | Isolati… — down | | GPU sharing used for sensitive workloads | Critical | Medium | Isolation profile, data-class restrictions, deny unknown sharing semantics | | Model/data locality ignored | Medium | High | Placement scoring incl…
| Reserved Cortex device requested by Cadence | Deny with `product_reservation`… — | Reserved Cortex device requested by Cadence | Deny with `product_reservation` | | Cloud burst without budget | Deny with `budget_required` | | Cloud burst without teardown proof path | Deny with `teardown_required` | …
ence. - Adapter compliance report. Steps: 1. Confirm sandbox authority and de… — ence. - Adapter compliance report. Steps: 1. Confirm sandbox authority and device approval. 2. Build a sandbox cluster or use an approved sandbox node pool. 3. Install only approved GPU components: - NVIDIA GPU Ope…
tify device class: physical host, VM, accelerator, model endpoint, storage, or … — tify device class: physical host, VM, accelerator, model endpoint, storage, or network service. 3. Gather read-only facts: OS, driver, GPU, VRAM, CPU, memory, storage, network, scheduler compatibility, metrics availabil…
x GPU job can run through UAPI. - Kueue queue/admission behavior is visible. - … — x GPU job can run through UAPI. - Kueue queue/admission behavior is visible. - GPU metrics are scraped. - Artifacts are stored with checksums. - No production/public exposure changes occurred. - GPU Operator/device-plug…
ded as versioned policy: - Who can request each lane. - What approvals are req… — ded as versioned policy: - Who can request each lane. - What approvals are required. - Which data classes may run where. - Which products have reservations. - Which devices are excluded. - Which cloud providers are all…
ples: - Artifact references should be immutable and checksummed. - Sensitive d… — ples: - Artifact references should be immutable and checksummed. - Sensitive datasets should require data-class policy and lane controls. - Model weights should have provenance, license, hash, quantization, runtime com…
del inference and evaluation jobs. - Use protected data-class and retention rul… — del inference and evaluation jobs. - Use protected data-class and retention rules. - Keep special GPU reservations explicit if they are still valid. Cortex should not: - Treat old reserved GPU assumptions as permanent…
oute/DNS/firewall/reverse proxy/NetBird mutation from compute events | Event su… — oute/DNS/firewall/reverse proxy/NetBird mutation from compute events | Event subscription or `/v1/runs/{run_id}` safe view | Notification/evidence cards | | Mesh | Service topology, readiness dependencies, cross-product…
0Z" preemption: preemptible: false checkpoint_ref: null requeue_policy: n… — 0Z" preemption: preemptible: false checkpoint_ref: null requeue_policy: never metrics: queue_wait_seconds: 30 runtime_seconds: 480 gpu_seconds: 480 peak_vram_gb: 17.4 cost_usd_estimate: 0.00 observabilit…
ema_version: uai.routing_policy.v0.1 policy_id: routing_policy_default_20260621… — ema_version: uai.routing_policy.v0.1 policy_id: routing_policy_default_20260621 status: draft applies_to: products: - cadence - cortex - beacon - mesh - explorer - command-center defaults: de…
PU vendor, model, VRAM, PCIe location, driver. - NVIDIA CUDA/driver, Intel Leve… — PU vendor, model, VRAM, PCIe location, driver. - NVIDIA CUDA/driver, Intel Level Zero/oneAPI/SYCL, ROCm, Metal, Vulkan as applicable. - Container runtime and GPU runtime. - Kubernetes/Slurm/Nomad/Proxmox compatibility. …
null network: lan_gbps: null reservations: [] observability: metrics_en… — null network: lan_gbps: null reservations: [] observability: metrics_endpoint: null dcgm_available: false otel_agent: false notes: - "Candidate record only. Do not route workloads until surveyed and approv…
roved`. - Data class: `public_safe`, `internal`, `private`, `secret_bearing`, `… — roved`. - Data class: `public_safe`, `internal`, `private`, `secret_bearing`, `customer`, `model_weights`, `dataset_sensitive`. - Workload type: `llm_inference`, `embedding`, `image_generation`, `batch_eval`, `training`…
s, private IPs, exact capacity, or reserved-lane details unless the caller has … — s, private IPs, exact capacity, or reserved-lane details unless the caller has a matching operator/product scope. Public-safe or Explorer-facing device summaries should use pseudonymous IDs, field allowlists, freshness …
ut never lower it. - Unknown artifact, model, dataset, or lineage class denies … — ut never lower it. - Unknown artifact, model, dataset, or lineage class denies by default. - `secrets_expected=true` is not permission to handle secrets; it is a routing warning that requires protected handling or denia…
write notes after the fact | | L1 | Typed request and run record | Products sub… — write notes after the fact | | L1 | Typed request and run record | Products submit structured requests and receive run IDs | | L2 | Registry-aware routing | Router filters by capabilities, reservations, health, and trus…
string device_type string owner string locality string trust_tier … — string device_type string owner string locality string trust_tier string exposure_boundary string schedulable_state } ACCELERATOR { string accelerator_id string vendor string model …
rect Host Lane\nmanual/special wrappers"] Cloud["Cloud GPU Burst Lane\nleas… — rect Host Lane\nmanual/special wrappers"] Cloud["Cloud GPU Burst Lane\nleased nodes with TTL/budget"] end subgraph Platform["Platform Services"] Models["Model Gateway\nLiteLLM, vLLM, Ollama"] Store["Art…
e satisfied | | `logs` | Return only redacted/product-authorized log views | | … — e satisfied | | `logs` | Return only redacted/product-authorized log views | | `artifacts` | Finalize output refs with checksums and retention policy | | `metrics` | Emit run labels, queue wait, runtime, GPU usage, cost…
d direct host or approved internal lane | | `sandbox` | Separate namespace/clus… — d direct host or approved internal lane | | `sandbox` | Separate namespace/cluster, low-trust policy, no secret mounts | Disposable Ray sandbox only | Separate test partition if used | Separate sandbox datacenter/client…
e | Containerized services, inference serving, repeatable jobs, quota-managed b… — e | Containerized services, inference serving, repeatable jobs, quota-managed batch with Kueue, gang jobs with Volcano, Ray clusters with KubeRay, cloud GPU node pools | Device inventory source of truth, Proxmox lifecyc…
nly hosts, Proxmox VMs, bare-metal Linux, and cloud GPU workers need different … — nly hosts, Proxmox VMs, bare-metal Linux, and cloud GPU workers need different runtime paths. - **Long-lived services versus batch jobs:** vLLM-style inference, Ollama local serving, embeddings, Ray distributed jobs, te…
such as Intel Arc Pro B70 to be first-class in the data model even before it is… — such as Intel Arc Pro B70 to be first-class in the data model even before it is operationally proven. - UAI wants priority lanes and smart routing, not a single FIFO queue. - UAPI exists conceptually as an internal brid…
ion only | `product_integration` approval | Per-product contract matrix and acc… — ion only | `product_integration` approval | Per-product contract matrix and acceptance evidence | Product bypasses UAPI or mutates forbidden surfaces | | Gate 7 - cloud burst | Use leased GPU capacity | Cloud resources …
r` | Backend-native allocator such as kube-scheduler, Kueue admission, Ray sche… — r` | Backend-native allocator such as kube-scheduler, Kueue admission, Ray scheduler, Slurm, or Nomad | | `runtime` | The actual workload runtime such as vLLM, Ollama, Ray, container job, shell wrapper, or benchmark run…
frastructure authority | Cadence requests; Command Center approves; UAPI enforc… — frastructure authority | Cadence requests; Command Center approves; UAPI enforces | | Client-declared data class is trusted | Products could accidentally lower sensitivity | Server-side classification resolver; unknown …
ts ask; UAPI admits and routes; backend schedulers allocate; GPU Cloud executes… — ts ask; UAPI admits and routes; backend schedulers allocate; GPU Cloud executes; Explorer records what is true.** ## How to read this proposal - **Operators and product owners:** read the executive thesis, decision pa…
uapi-redesign) 13. [Smart routing and priority lane design](#smart-routing-and-… — uapi-redesign) 13. [Smart routing and priority lane design](#smart-routing-and-priority-lane-design) 14. [Device registry design](#device-registry-design) 15. [Workload request schema](#workload-request-schema) 16. [Rou…
# UAI GPU Cloud + UAPI Bridge Lab Ecosystem Redesign Metadata: - Status: curre… — # UAI GPU Cloud + UAPI Bridge Lab Ecosystem Redesign Metadata: - Status: current - Product: UAI GPU Cloud + UAPI Bridge - Source Project: UAI Thinktank - Created: 2026-06-23T18:49:59.150538+00:00 - Provenance: UAI Thin…
UAI Thinktank Explorer visibility check remains blocked after promotion. Promot… — UAI Thinktank Explorer visibility check remains blocked after promotion. Promoted slug uai-doc-uai-thinktank-ai-context returned context status 403 to the current AI lane. The promoted item requires the uai-thinktank ta…
UAI Thinktank Explorer publication intake packet is ready for review. Product:… — UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
UAI Thinktank Explorer visibility check remains blocked after promotion. Promot… — UAI Thinktank Explorer visibility check remains blocked after promotion. Promoted slug uai-doc-uai-thinktank-ai-context returned context status 403 to the current AI lane. The promoted item requires the uai-thinktank ta…
UAI Thinktank Explorer publication intake packet is ready for review. Product:… — UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
UAI Thinktank Explorer visibility check remains blocked after promotion. Promot… — UAI Thinktank Explorer visibility check remains blocked after promotion. Promoted slug uai-doc-uai-thinktank-ai-context returned context status 403 to the current AI lane. The promoted item requires the uai-thinktank ta…
Thinktank bridge P1: /operator/ideas thinktank lane + promotion concept published v0.7.3. Tag ACL proposal only — no mutation.
Fable curation pass 2026-07-09 (post-mesh pivot). Scanned 343 AI-visible catalog entries, read 27 context packets, ranked 5 early-stage project picks. Recommended first: ecosystem capability card repair (ghost product-c…
Delivered lightweight OpenAI-compatible supergrok relay service for lab/Thinktank use. Alpine container. Supports specialty agents (planner, coder, thinktank, uai). Includes groupchat.py for multi-agent orchestration. F…
Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canonical `uai-thinktank`-tagged context remains forbidden to this AI lane. This is documentation visibility only and grants no runtime, provider…
UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
Create an AI-readable Explorer runbook bridge for UAI Thinktank while the canonical `uai-thinktank`-tagged context remains forbidden to this AI lane. This is documentation visibility only and grants no runtime, provider…
UAI Thinktank Explorer visibility check remains blocked after promotion. Promoted slug uai-doc-uai-thinktank-ai-context returned context status 403 to the current AI lane. The promoted item requires the uai-thinktank ta…
UAI Thinktank Explorer publication intake packet is ready for review. Product: uai-thinktank Task: catalog/review the local v0.1 operator-review packet in UAI Explorer. Authority: documentation/catalog intake only. Pub…
UAI GPU Cloud + UAPI Bridge plan ingest reached Explorer feedback and promotion, but the promoted context is not readable by this AI lane. Target slug: [REDACTED:high-entropy]. Promotion request id: 196. Catalog item id…
proved sidecars can paste bounded Codex, GLM, Grok, browser-relay, or local-worker critique outputs into the External Result Intake card, but the proposal must still treat those outputs as review evidence, not imp…
comparison sources without ranking official docs versus anecdotal/community evidence. That matters for Intel Arc Pro B70 and scheduler maturity. - The embedded self-review appendix is valuable but awkward. It criticizes…
x. Some of those critiques are sharper than the main body, especially on schema incompleteness, storage pressure, and K8s-on-Proxmox risk. What I do not like: - The opening "Connection TL;DR" is a serious hygiene fail…
remote artifact storage: https://mlflow.[REDACTED:high-entropy]/ - MLflow artifact stores: https://mlflow.[REDACTED:high-entropy]/ - MinIO AIStor container/S3-family docs: https://docs.min.io/aistor/installation/contain…
https://docs.vllm.ai/ - vLLM OpenAI-compatible server: https://docs.vllm.[REDACTED:high-entropy].html - vLLM Kubernetes/Helm deployment docs: https://docs.vllm.ai/en/stable/deployment/frameworks/helm.html - Ollama API d…
Runbook, fetched through AI context slug `[REDACTED:high-entropy]`. - UAI Development Standards Runbook, fetched through AI context slug `[REDACTED:high-entropy]`. - UAI Agent Feedback Convention, fetched through AI con…
ng | LiteLLM belongs in the model-provider/gateway layer, not the hardware scheduler; budgets should be mirrored in UAPI policy. | | OpenTelemetry | The OpenTelemetry Collector is a vendor-agnostic receiver/processor/ex…
nit of admission for applications that run to completion, and AdmissionChecks allow extra criteria before admission. Sources: https://kueue.sigs.k8s.io/docs/concepts/workload/ and https://kueue.sigs.k8s.io/docs/concepts…
down | | GPU sharing used for sensitive workloads | Critical | Medium | Isolation profile, data-class restrictions, deny unknown sharing semantics | | Model/data locality ignored | Medium | High | Placement scoring incl…
| Reserved Cortex device requested by Cadence | Deny with `product_reservation` | | Cloud burst without budget | Deny with `budget_required` | | Cloud burst without teardown proof path | Deny with `teardown_required` | …
ence. - Adapter compliance report. Steps: 1. Confirm sandbox authority and device approval. 2. Build a sandbox cluster or use an approved sandbox node pool. 3. Install only approved GPU components: - NVIDIA GPU Ope…
tify device class: physical host, VM, accelerator, model endpoint, storage, or network service. 3. Gather read-only facts: OS, driver, GPU, VRAM, CPU, memory, storage, network, scheduler compatibility, metrics availabil…
x GPU job can run through UAPI. - Kueue queue/admission behavior is visible. - GPU metrics are scraped. - Artifacts are stored with checksums. - No production/public exposure changes occurred. - GPU Operator/device-plug…
ded as versioned policy: - Who can request each lane. - What approvals are required. - Which data classes may run where. - Which products have reservations. - Which devices are excluded. - Which cloud providers are all…
ples: - Artifact references should be immutable and checksummed. - Sensitive datasets should require data-class policy and lane controls. - Model weights should have provenance, license, hash, quantization, runtime com…
del inference and evaluation jobs. - Use protected data-class and retention rules. - Keep special GPU reservations explicit if they are still valid. Cortex should not: - Treat old reserved GPU assumptions as permanent…
oute/DNS/firewall/reverse proxy/NetBird mutation from compute events | Event subscription or `/v1/runs/{run_id}` safe view | Notification/evidence cards | | Mesh | Service topology, readiness dependencies, cross-product…
0Z" preemption: preemptible: false checkpoint_ref: null requeue_policy: never metrics: queue_wait_seconds: 30 runtime_seconds: 480 gpu_seconds: 480 peak_vram_gb: 17.4 cost_usd_estimate: 0.00 observabilit…
ema_version: uai.routing_policy.v0.1 policy_id: routing_policy_default_20260621 status: draft applies_to: products: - cadence - cortex - beacon - mesh - explorer - command-center defaults: de…
PU vendor, model, VRAM, PCIe location, driver. - NVIDIA CUDA/driver, Intel Level Zero/oneAPI/SYCL, ROCm, Metal, Vulkan as applicable. - Container runtime and GPU runtime. - Kubernetes/Slurm/Nomad/Proxmox compatibility. …
null network: lan_gbps: null reservations: [] observability: metrics_endpoint: null dcgm_available: false otel_agent: false notes: - "Candidate record only. Do not route workloads until surveyed and approv…
roved`. - Data class: `public_safe`, `internal`, `private`, `secret_bearing`, `customer`, `model_weights`, `dataset_sensitive`. - Workload type: `llm_inference`, `embedding`, `image_generation`, `batch_eval`, `training`…
s, private IPs, exact capacity, or reserved-lane details unless the caller has a matching operator/product scope. Public-safe or Explorer-facing device summaries should use pseudonymous IDs, field allowlists, freshness …
ut never lower it. - Unknown artifact, model, dataset, or lineage class denies by default. - `secrets_expected=true` is not permission to handle secrets; it is a routing warning that requires protected handling or denia…
write notes after the fact | | L1 | Typed request and run record | Products submit structured requests and receive run IDs | | L2 | Registry-aware routing | Router filters by capabilities, reservations, health, and trus…
string device_type string owner string locality string trust_tier string exposure_boundary string schedulable_state } ACCELERATOR { string accelerator_id string vendor string model …
rect Host Lane\nmanual/special wrappers"] Cloud["Cloud GPU Burst Lane\nleased nodes with TTL/budget"] end subgraph Platform["Platform Services"] Models["Model Gateway\nLiteLLM, vLLM, Ollama"] Store["Art…
e satisfied | | `logs` | Return only redacted/product-authorized log views | | `artifacts` | Finalize output refs with checksums and retention policy | | `metrics` | Emit run labels, queue wait, runtime, GPU usage, cost…
d direct host or approved internal lane | | `sandbox` | Separate namespace/cluster, low-trust policy, no secret mounts | Disposable Ray sandbox only | Separate test partition if used | Separate sandbox datacenter/client…
e | Containerized services, inference serving, repeatable jobs, quota-managed batch with Kueue, gang jobs with Volcano, Ray clusters with KubeRay, cloud GPU node pools | Device inventory source of truth, Proxmox lifecyc…
nly hosts, Proxmox VMs, bare-metal Linux, and cloud GPU workers need different runtime paths. - **Long-lived services versus batch jobs:** vLLM-style inference, Ollama local serving, embeddings, Ray distributed jobs, te…
2026-07-12 naming: Diffuser=internal ecosystem router; UAPI=future public embed wing. Published: [REDACTED:high-entropy], [REDACTED:high-entropy], [REDACTED:high-entropy]. Please amend concise infra map [REDACTED:high-e…
Pass1: 8 St Johns/ATB weak-web candidates documented. Pass2: Trammel+Thaladex flagship framing; STU/Skipper Thaladex-backed. Pass3: preview-only at https://www.united-ai.ca/preview/ — dream-it, composites Lemarchant/Top…
2026-07-11: Owner GO checklist authored. Verdict NO-GO (known FAIL public mail 502). Restore runbook not safe-auto; no restore started. Catalog evidence: [REDACTED:high-entropy]. State+restore See Also synced. Mac archi…
Owner Decision 1 = A (2026-07-11): Explorer lab write-SOT lives on VPS/beacon after Director rebuild — not Echo, not Director. Public Explorer already on beacon :8081 / explorer.united-ai.ca v0.7.6 (public_safe_mode). T…
Director posture 2026-07-11: Director may need Proxmox reinstall; post-rebuild role = bulk storage + KEEP sandboxes (CT123/124/126) — NOT primary Explorer write-SOT / NOT latency-sensitive control plane. CT121 lab write…
Operator correction (2026-07-09): VPS loopback 127.0.0.1:8081 is uapi-explorer.service (UAI Explorer public catalog v0.7.0), NOT a cloud-GPU broker. Beacon proxies explorer.united-ai.ca to this port. Prior operator rece…
2026-07-05 human docs reorg audit (owner feedback: documentation mess, jumbled UX). Top pain points for human navigation: 1) /docs renders flat 407-row wall (~310KB) with no default filter — nav label "Runbooks" mislea…
2026-07-04 wipe session documentation pass COMPLETE. Handoffs index: [REDACTED:high-entropy].md. Session CLOSED: Stage 2 VPS DONE, Stage 2.5 Proxmox reboot (Echo OK, Director DOWN), Cadence Fable DEPLOYED, START_HERE + …
Governance vNext now includes no-local-copies / local_data_hoarding=never rule (2026-07-04). Consider surfacing in bootstrap required_rules or start-here MOC card for agent visibility.
Trammel preserve published: product + index; preserved_partial; Beacon inactive route retained. Gaps: VPS static roots missing pre-rescue.
Observation: operator starting new product Pipfitter — iOS smart spirit level leveraging phone sensors, Stabila-inspired UX (audio guidance, REF mode, mm/m and degree readouts). Research phase complete; no catalog entry…
Mac-side clean-slate-2026-07-01 handoff documents are not synced to Explorer. During a 2026-07-02 long-pass session on the VPS (/root/uapi-service-lab), the operator-referenced Mac paths (/Users/scott/Documents/New [RED…
Codex B70 preinstall update: Forge is confirmed as target node 10.0.0.248; Echo is 10.0.0.249; CT103 Engineering is 10.0.0.103. HomeLab was read-only checked as quorate 3 nodes with HA idle at observation time. Forge cu…
UAI GPU Cloud + UAPI Bridge plan ingest reached Explorer feedback and promotion, but the promoted context is not readable by this AI lane. Target slug: [REDACTED:high-entropy]. Promotion request id: 196. Catalog item id…
Wave 13 receipt promotion verification issue. Promotion request returned status=200, [REDACTED:high-entropy], feedback_id=277, promotion_audit=2069. Catalog matches=0 and context_status=403. Please repair AI-visible cat…
# UAPI Wave 13 Cloud/Cortex/GLM Agent-Team Receipt Authority: fixture_only_no_execution. No service activation, dispatch, backend calls, storage writes, memory writes, ledger writes, SSH, mounts, Colab/cloud calls, or …
Wave 12 receipt promotion verification issue. Promotion request returned status=200, [REDACTED:high-entropy], feedback_id=275, promotion_audit=2049. Catalog matches=0 and context_status=403. Please repair AI-visible cat…
# UAPI Wave 12 Route/Forge/GLM Agent-Team Receipt Authority: fixture_only_no_execution. No service activation, dispatch, backend calls, storage writes, ledger writes, SSH, mounts, Colab/cloud calls, or route activation…
Wave 11 receipt promotion verification issue. Promotion request returned status=200, [REDACTED:high-entropy], feedback_id=273, promotion_audit=2029. Catalog matches=0 and context_status=403. Please repair AI-visible cat…
# UAPI Wave 11 Forge/LabGPU Agent-Team Receipt Authority: fixture_only_no_execution. No service activation, dispatch, backend calls, ledger writes, SSH, mounts, Colab/cloud calls, or route activation. Commit: fb3d348 …
Repair request: UAPI Wave 10 route/placement/autorouter evidence receipt promotion returned status=promoted with audit_event_id=2001 from source feedback id=271 audit_event_id=2000, but GET /api/ai/catalog has 0 matches…
UAPI Wave 10 merged receipt. Commit e5080fe. Grok Builds lanes used model=grok-build effort=max. Wave 10 A attached NUC benchmark evidence refs to local.nuc.gemma.npu and local.nuc.gemma.140t smoke benchmark gates while…
Repair request: UAPI Wave 9 NUC and Autorouter merge receipt promotion returned status=promoted with audit_event_id=1979 from source feedback id=269 audit_event_id=1978, but GET /api/ai/catalog has 0 matches for target …
UAPI Wave 9 merged receipt. Commit 96fa654. Grok Builds lanes used model=grok-build effort=max. GLM 5.2 lane attempted with --effort max and failed with provider 529, recorded for retry. Implemented fixture-only NUC acc…
Explorer repair request for promoted UAPI cloud burst teardown evidence receipt. Receipt feedback id 267 audit 1957 was submitted for target slug [REDACTED:high-entropy]. Promotion returned status promoted with audit 1…
UAPI cloud-burst teardown evidence receipt. Scope: - Added fixture-only cloud burst teardown evidence packet: fixtures/cloud-burst-teardown-evidence.v0.json. - Added validator module: [REDACTED:high-entropy].py. - Adde…
Explorer repair request for promoted UAPI storage capability API contract receipt. Receipt feedback id 265 audit 1952 was submitted for target slug [REDACTED:high-entropy]. Promotion returned status promoted with audit…
UAPI storage capability API contract receipt. Scope: - Added fixture-only storage capability API packet: fixtures/storage-capability-api.v0.json. - Added read-only API contract validator/report module: src/uapi_service…
Explorer repair request for promoted UAPI Wave 8 agent team queue receipt. Receipt feedback id 263 audit 1939 was submitted for target slug [REDACTED:high-entropy]. Promotion returned status promoted with audit 1940. …
UAPI Wave 8 agent team queue receipt. Created queue/control files for three Grok Builds lanes and one GLM 5.2 lane: - Grok Wave 8 A: storage capability API contract. - Grok Wave 8 B: NUC 140T/NPU accelerator next bench…
Explorer repair request for promoted UAPI storage capability evidence receipt. Receipt feedback id 261 audit 1896 was submitted for target slug [REDACTED:high-entropy]. Promotion returned status promoted with audit 189…
UAPI storage capability evidence receipt for the Forge SN850X/UAPI-served storage wedge. Scope: - Added fixture-only StorageCapabilityPacket as fixtures/storage-capability.v0.json. - Added dependency-free validator mod…
# Explorer Publication Repair Request - UAPI Dynamic Route Alias Denial Evidence Receipt Target slug: [REDACTED:high-entropy] Feedback id: 259 Feedback audit: 1890 Promotion audit: 1891 Promotion status: promoted Catal…
# UAPI Dynamic Route Alias Denial Evidence Receipt - 2026-06-23 Authority: fixture-only, no execution, no service activation. Product: uapi Kind: runbook receipt Commit: 262b42d Implementation commit: cd359ce Summary:…
# Explorer Publication Repair Request - UAPI NUC Accelerator Benchmark Evidence Hardening Receipt Target slug: [REDACTED:high-entropy] Feedback id: 257 Feedback audit: 1884 Promotion audit: 1885 Promotion status: promo…
# UAPI NUC Accelerator Benchmark Evidence Hardening Receipt - 2026-06-23 Authority: fixture-only, no execution, no service activation. Product: uapi Kind: runbook receipt Commit: b5c1fcc Implementation commit: f0c8389 …
# Explorer Publication Repair Request - UAPI Forge Storage Inventory Evidence Receipt Target slug: [REDACTED:high-entropy] Feedback id: 255 Feedback audit: 1856 Promotion audit: 1857 Promotion status: promoted Catalog …
# UAPI Forge Storage Inventory Evidence Receipt - 2026-06-23 Authority: fixture-only, no execution, no service activation. Product: uapi Kind: runbook receipt Commit: f6f184e Implementation commit: e90284c Summary: Co…
Explorer promotion/readback issue for [REDACTED:high-entropy]. Feedback id 253 audit 1850 was submitted and promotion requested, but catalog_matches=0 and context_status=403. Please repair AI-lane catalog/context visibi…
# UAPI Placement Friendly Route Denial Evidence Receipt Authority: fixture_only_no_execution. No service activation, no backend dispatch, no hardware execution, no route activation. Commit: 55febcd Add placement frien…
Explorer promotion/readback issue for [REDACTED:high-entropy]. Feedback id 250 audit 1842 was submitted and promotion requested, but catalog_matches=0 and context_status=403. Please repair AI-lane catalog/context visibi…
# UAPI Structured Benchmark No-Fallback Proof Receipt Authority: fixture_only_no_execution. No service activation, no backend dispatch, no hardware execution, no route activation. Commit: e42b674 Require structured be…
Explorer promotion/readback issue for [REDACTED:high-entropy]. Feedback id audit was submitted and promotion requested, but catalog_matches=0 and context_status=404. Please repair AI-lane catalog/context visibility or…
Target slug [REDACTED:high-entropy] was submitted via feedback id 247 and promotion request, but catalog matches=0 and context status=403. Please review promotion/import permissions so AI-lane receipts promoted from fee…
Target slug [REDACTED:high-entropy] was submitted via feedback id 245 and promotion request, but catalog matches=0 and context status=403. Please review promotion/import permissions so AI-lane receipts promoted from fee…
Target slug [REDACTED:high-entropy] was submitted via feedback id 243 and promotion request, but catalog matches=0 and context status=403. Please review promotion/import permissions so AI-lane receipts promoted from fee…
Explorer publication repair request: UAPI receipt promotion returned status=promoted and next_step=verify_context_fetch for slug [REDACTED:high-entropy], but AI-lane verification failed. /api/ai/catalog returned 200 wit…
Explorer publication repair request: UAPI receipt promotion returned status=promoted and next_step=verify_context_fetch for slug [REDACTED:high-entropy], but AI-lane verification failed. /api/ai/catalog returned 200 wit…
Explorer publication repair request: UAPI receipt promotion returned status=promoted and next_step=verify_context_fetch for slug [REDACTED:high-entropy], but AI-lane verification failed. /api/ai/catalog returned 200 wit…
Explorer publication repair request: UAPI receipt promotion returned status=promoted and next_step=verify_context_fetch for slug [REDACTED:high-entropy], but AI-lane verification failed. /api/ai/catalog returned 200 wit…
Explorer promotion/readback issue for UAPI Wave 5 receipt. Target slug: [REDACTED:high-entropy] Feedback id: 233 Promotion audit_event_id: 1783 Context readback status: 403 Please repair AI-lane visibility or explain re…
Explorer promotion/readback issue for UAPI Wave 5 receipt. Target slug: [REDACTED:high-entropy] Feedback id: 231 Promotion audit_event_id: 1779 Context readback status: 403 Please repair AI-lane visibility or explain re…
Explorer promotion/readback issue for UAPI Wave 4 GLM retry receipt. Target slug: [REDACTED:high-entropy] Feedback id: 229 Promotion audit_event_id: 1772 Context readback status: 403 Please repair AI-lane visibility or …
Explorer promotion/readback issue for UAPI provider-error receipt. Target slug: [REDACTED:high-entropy] Feedback id: 227 Promotion audit_event_id: 1766 Promotion endpoint returned status=promoted, but GET /api/ai/contex…
# UAI Explorer Promotion Readback Issue Promotion returned promoted for public safe UAPI receipt feedback, but the AI lane could not read the promoted context. Observed sequence: - Ready was healthy with public safe m…
Promotion/readback repair request for UAPI benchmark harness wedge receipt. The AI lane accepted feedback id 208 for target slug [REDACTED:high-entropy] and accepted a promotion request with audit_event_id 1673. The pr…
Long-pass agent-team summaries feedback ids 197-200 promoted but are not AI-visible: direct /api/ai/context fetch returns 403 not authorized and /api/ai/catalog has no matching longpass slugs. A repair promotion was att…
# UAPI Seventeenth-Wedge Provider Adapter Receipt (2026-06-22) Created: 2026-06-23T00:42:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Ad…
# UAPI Sixteenth-Wedge Policy Bundle Receipt (2026-06-22) Created: 2026-06-23T00:26:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added …
# UAPI Fifteenth-Wedge Event Lifecycle Receipt (2026-06-22) Created: 2026-06-23T00:12:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added…
# UAPI Fourteenth-Wedge Route Activation Gates Receipt (2026-06-22) Created: 2026-06-22T23:58:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` …
# UAPI Thirteenth-Wedge GitNexus Index Receipt (2026-06-22) Created: 2026-06-22T23:45:00Z Authority: local repository intelligence receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Git…
# UAPI Twelfth-Wedge CI Metadata Receipt (2026-06-22) Created: 2026-06-22T23:28:00Z Authority: local fixture-only scaffold receipt. No CI run registered. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added - `.g…
# UAPI Eleventh-Wedge Fixture Check Script Receipt (2026-06-22) Created: 2026-06-22T23:02:00Z Authority: local fixture-only scaffold receipt. No automation installed. ## Target Repo: - `/[REDACTED:high-entropy]` ##…
# UAPI Tenth-Wedge Acceptance Summary CLI Receipt (2026-06-22) Created: 2026-06-22T22:42:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Ad…
# UAPI Ninth-Wedge Redacted Evidence Packet Receipt (2026-06-22) Created: 2026-06-22T22:24:00Z Authority: local fixture-only scaffold receipt. No Explorer export from service. ## Target Repo: - `/[REDACTED:high-entr…
# UAPI Eighth-Wedge RunRecord Sink Receipt (2026-06-22) Created: 2026-06-22T22:10:00Z Authority: local fixture-only scaffold receipt. No durable ledger. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added - `src…
# UAPI Seventh-Wedge JobBrief Receipt (2026-06-22) Created: 2026-06-22T21:55:00Z Authority: local fixture-only scaffold receipt. No execution route. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added - `src/uap…
# UAPI Sixth-Wedge Stored Session Preflight Receipt (2026-06-22) Created: 2026-06-22T21:38:00Z Authority: local fixture-only scaffold receipt. No real auth integration. ## Target Repo: - `/[REDACTED:high-entropy]` …
# UAPI Fifth-Wedge Fixture Session Store Receipt (2026-06-22) Created: 2026-06-22T21:24:00Z Authority: local fixture-only scaffold receipt. No real auth integration. ## Target Repo: - `/[REDACTED:high-entropy]` ## …
# UAPI Fourth-Wedge Route Registry and Facade Receipt (2026-06-22) Created: 2026-06-22T21:05:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` #…
# UAPI Codex Operator Rollup First Wedge (2026-06-22) Authority: Codex acting-operator rollup. No live service, hardware, cloud, storage, or route activation. ## Published Contexts This Loop - `[REDACTED:high-entropy…
# UAPI Third-Wedge Validator CLI Receipt (2026-06-22) Created: 2026-06-22T20:35:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added - `s…
# UAPI Second-Wedge Fixture Import Receipt (2026-06-22) Created: 2026-06-22T20:28:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Repo: - `/[REDACTED:high-entropy]` ## Added - …
# UAPI First-Wedge Scaffold Receipt (2026-06-22) Created: 2026-06-22T20:18:00Z Authority: local fixture-only scaffold receipt. No service activation. ## Target Created new isolated local git repo: - `/[REDACTED:high…
# GLM 5.2 UAPI First-Wedge Test Plan (2026-06-22) Created: 2026-06-22T20:10:00Z Authority: GLM Worker F. Implementation-agnostic; portable once Codex chooses the repo. ## Scope Tests validate v0.2 contracts against f…
# GLM 5.2 UAPI Scaffold Target Gate (2026-06-22) Created: 2026-06-22T20:10:00Z Authority: GLM Worker F scaffold evaluation. No code edits, no repo mutations. ## Target Options Compared ### A: New Isolated UAPI Repo -…
# Grok UAPI Lab Route Readiness Map (2026-06-22) **ALIVE:** true **Lane:** Grok Worker F **Authority:** read-only. Candidate names only. No activation/writes/benches. ## Sources Required: forge-sn850x-storage-policy-g…
# Grok UAPI Implementation Target Map (2026-06-22) **ALIVE:** true **Lane:** Grok Worker F **Authority:** read-only. No code/service changes. ## Sources Required: uai-agent-dispatch diaries (uapi-implementation-target…
# UAPI Implementation Target Discovery (2026-06-22) Created: 2026-06-22T19:59:37Z Authority: read-only discovery evidence. No implementation edits. ## Question Find the current production UAPI implementation target b…