tao-run-automl-deft-pipeline

CID baeareifvwpscsvsj63rfqglp34zgqs22f6ediolymaizkg4uee3urdu7ca
v0.1.0 OASF 1.1.0 Not signed

Security

CID matches content
Yes
Signed
No
0 signature(s)
Trusted (signature verified)
No
pull signatures: pull referrer baeareidnhxtd2eaflvzcgg4lc4juvuu3fr3qwks5baeri6747oecvhgpiq: empty referrer baeareidnhxtd2eaflvzcgg4lc4juvuu3fr3qwks5baeri6747oecvhgpiq
OASF record valid
Yes

Record

Authors
NVIDIA
Created
2026-09-14T00:59:26Z
Pulled
1 day, 2 hours ago
Modules
core/language_model/agentskills
Raw OASF record (JSON)
{
  "authors": [
    "NVIDIA"
  ],
  "created_at": "2026-09-14T00:59:26Z",
  "description": "Run the canonical NVIDIA AOI three-phase training pipeline \u2014 Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA \u2192 SDG \u2192 mining \u2192 plain-train retrain), Phase 3 AutoML refinement on the DEFT-augmented dataset. Use when the user asks to \"run the AOI workflow\", \"fine-tune my PCB AOI model end-to-end\", \"improve my AOI ChangeNet model\", or \"AOI workflow with AutoML\" request \u2014 route here instead of tao-run-deft-aoi directly unless the user explicitly asks for the DEFT loop ONLY (e.g. \"run JUST the DEFT loop\", \"skip AutoML, only DEFT\"). Also handles the same three-phase pattern for non-AOI DEFT applications \u2014 AutoML baseline then DEFT loop warm-started from AutoML's winning HPs then post-DEFT AutoML refinement on the iteration-augmented dataset. Trigger phrases include \"run the AOI workflow\", \"AOI end-to-end\", \"AutoML + DEFT\", \"AutoML then DEFT\", \"tune hyperparameters then DEFT\", \"DEFT with AutoML at both ends\", \"warm-start DEFT\", \"improve my AOI model\".",
  "domains": [
    {
      "id": 601,
      "name": "education/e_learning"
    },
    {
      "id": 503,
      "name": "human_resources/training_and_development"
    },
    {
      "id": 2302,
      "name": "insurance/claims_processing"
    },
    {
      "id": 205,
      "name": "finance_and_business/retail"
    },
    {
      "id": 11104,
      "name": "technology/artificial_intelligence/large_language_models"
    }
  ],
  "modules": [
    {
      "artifact": {
        "data": "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",
        "digest": "sha256:3856c12326697bc062c8f877b165e571ed240cab58e11906760e0f5518dbcd1a",
        "media_type": "application/agent-skills+gzip",
        "size": 20348
      },
      "data": {
        "artifacts": [
          {
            "artifact_hash": "sha256:9ca6bb86cf2ba049b9f5194a34f5f635ce942b21f8a8f39fb69e4a2054e86eb2",
            "path": "BENCHMARK.md",
            "type": "other"
          },
          {
            "artifact_hash": "sha256:16ce7215be9a628ab45c29511699b247c8269d86e0a9e9ee46a72dd82f8a1c17",
            "path": "evals/evals.json",
            "type": "other"
          },
          {
            "artifact_hash": "sha256:aa6284eae52d187c3f4f6ad1b408eab20a7a7665fda069c7c035fdb6fc0e7bc9",
            "path": "references/consolidated-preflight.md",
            "type": "reference"
          },
          {
            "artifact_hash": "sha256:78861d45e1eaf349b37296a11e6df923883ae46b3da953120eb8120d34bed6bd",
            "path": "references/phase-handoffs.md",
            "type": "reference"
          },
          {
            "artifact_hash": "sha256:bbd911103fbb67f51e21d25493e55578e180dedf7f35880b18deb4c6ec6101ad",
            "path": "references/pitfalls-and-quality-checks.md",
            "type": "reference"
          },
          {
            "artifact_hash": "sha256:545a6429deaa64286735b2b187d7f8215d829add132d952f6641a4b62b8756bb",
            "path": "references/quick-start-example.md",
            "type": "reference"
          },
          {
            "artifact_hash": "sha256:271b4e9ba51fb3cf6b616e775bb3febafce8a064e4b1ffa0d21a21c4864da4b0",
            "path": "skill-card.md",
            "type": "other"
          },
          {
            "artifact_hash": "sha256:42886da387789af31c57d49d83ff92b1381ea37f0b8e1106cb7a7ffb92538332",
            "path": "skill.oms.sig",
            "type": "other"
          }
        ],
        "skill_file": "SKILL.md",
        "skill_manifest": {
          "allowed_tools": [
            "Read",
            "Skill",
            "Bash",
            "Write"
          ],
          "compatibility": [
            "Requires docker + nvidia-container-toolkit. Workflows (tao-run-automl, tao-run-deft-aoi) declare additional requirements."
          ],
          "description": "Run the canonical NVIDIA AOI three-phase training pipeline \u2014 Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA \u2192 SDG \u2192 mining \u2192 plain-train retrain), Phase 3 AutoML refinement on the DEFT-augmented dataset. Use when the user asks to \"run the AOI workflow\", \"fine-tune my PCB AOI model end-to-end\", \"improve my AOI ChangeNet model\", or \"AOI workflow with AutoML\" request \u2014 route here instead of tao-run-deft-aoi directly unless the user explicitly asks for the DEFT loop ONLY (e.g. \"run JUST the DEFT loop\", \"skip AutoML, only DEFT\"). Also handles the same three-phase pattern for non-AOI DEFT applications \u2014 AutoML baseline then DEFT loop warm-started from AutoML's winning HPs then post-DEFT AutoML refinement on the iteration-augmented dataset. Trigger phrases include \"run the AOI workflow\", \"AOI end-to-end\", \"AutoML + DEFT\", \"AutoML then DEFT\", \"tune hyperparameters then DEFT\", \"DEFT with AutoML at both ends\", \"warm-start DEFT\", \"improve my AOI model\".",
          "frontmatter_metadata": {
            "author": "NVIDIA Corporation",
            "version": "0.1.0"
          },
          "license": "Apache-2.0",
          "name": "tao-run-automl-deft-pipeline",
          "version": "0.1.0"
        }
      },
      "id": 10302,
      "name": "core/language_model/agentskills"
    }
  ],
  "name": "tao-run-automl-deft-pipeline",
  "schema_version": "1.1.0",
  "skills": [
    {
      "id": 70203,
      "name": "ai_ml_engineering/model_training/model_fine_tuning"
    },
    {
      "id": 702,
      "name": "ai_ml_engineering/model_training"
    }
  ],
  "version": "0.1.0"
}

Advertisements

GossipSub broadcast

First seen
1 month ago
Last seen
38 minutes ago
Announcements
1,127
Publishers
1
Topics
dir/labels/v1
Every 2 hours, last 3 days
Announcements Publishers
Announcements 30
Nov 14 Nov 15 Nov 16
10 Publishers

DHT add-provider

DHT key
QmaZwbjDAvXTD5a3gv2J5HTVrMcLJzo65XvnDXYdJSB1pj
First seen
6 days, 23 hours ago
Last seen
38 minutes ago
Requests
155
Providers
1
Lookups
6
Seekers
6
Last lookup
1 day, 2 hours ago
Every 2 hours, last 3 days
ADD_PROVIDER GET_PROVIDERS Providers
Requests 30
Nov 14 Nov 15 Nov 16
10 Providers

Skills

  • ai_ml_engineering 1
    • model_training 1
      • model_fine_tuning

Domains

  • education 1
    • e_learning
  • finance_and_business 1
    • retail
  • human_resources 1
    • training_and_development
  • insurance 1
    • claims_processing
  • technology 1
    • artificial_intelligence 1
      • large_language_models

Locators

None

Annotations

oasf-version
1.1.0
name
tao-run-automl-deft-pipeline
version
0.1.0

GossipSub publishers

DHT providers

github.com/libp2p/go-libp2p Expires in 1 day, 23 hours
Requests
155
First seen
6 days, 23 hours ago
Last seen
38 minutes ago
Connection address
/ip4/3.140.179.223/udp/57133/quic-v1
Listen addresses (3)
  • /ip4/10.0.17.42/tcp/5555
  • /ip4/127.0.0.1/tcp/5555
  • /ip4/3.150.168.105/tcp/5555
Protocols (14)
  • /dir/rpc/1.0.0
  • /floodsub/1.0.0
  • /ipfs/id/1.0.0
  • /ipfs/id/push/1.0.0
  • /ipfs/ping/1.0.0
  • /libp2p/autonat/1.0.0
  • /libp2p/circuit/relay/0.2.0/hop
  • /libp2p/circuit/relay/0.2.0/stop
  • /libp2p/dcutr
  • /meshsub/1.0.0
  • /meshsub/1.1.0
  • /meshsub/1.2.0
  • /meshsub/1.3.0
  • dir/kad/1.0.0

Latest pull attempts

Time Provider Providers found Lookup Connect Pull Total CID Signatures Trusted OASF Result
1 day, 2 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 6,363 ms 0 ms 4,655 ms 11,019 ms 0 OK
2 days, 3 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 6,929 ms 0 ms 5,083 ms 12,013 ms 0 OK
3 days, 6 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 5,304 ms 0 ms 7,011 ms 12,316 ms 0 OK
4 days, 8 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 5,012 ms 0 ms 4,011 ms 9,023 ms 0 OK
5 days, 9 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 6,168 ms 0 ms 8,903 ms 15,072 ms 0 OK
6 days, 12 hours ago 12D3KooWLf9p3cedc86xGQBaqak6rAFmQk1HxKAK1yh7umHE3amu 1 10,000 ms 0 ms 6,229 ms 16,229 ms 1 OK