opensite-ai/opensite-skills / aws-ai-ml
aws-ai-ml
VerifiedSelects, deploys, and customizes AI models on Amazon SageMaker. Training or Processing jobs, fine-tuning (SFT/DPO/RLVR/RLAIF), model selection, dataset preparation, evaluation, SageMaker or Bedrock deployment, inference optimization and endpoint diagnostics. Covers the full lifecycle from planning through production. Use when fine-tuning models on SageMaker, choosing which base model to customize, fine-tune, or deploy from SageMaker JumpStart or Hub, SageMakerPublicHub, or the SageMaker public model catalog, transforming or validating training data, evaluating model quality, deploying or optimizing endpoints, configuring IAM/S3 for training, or managing SageMaker Managed MLflow. Use for endpoint health, failures, latency, logs, metrics, errors. Covers Serverless Model Customization, Nova and OSS deployment paths, and PySDK v3. NOT for Ground Truth labeling, Feature Store, or general-purpose AWS infrastructure.
references/processing-jobs/references/troubleshooting.md
View on GitHub# Processing Job Troubleshooting
## Collect evidence
1. Run `aws sagemaker describe-processing-job --processing-job-name <name> --region <region>`.
2. Record `ProcessingJobStatus`, `ExitMessage`, `FailureReason`, `AppSpecification`, `ProcessingResources`, `ProcessingInputs`, `ProcessingOutputConfig`, and `RoleArn`.
3. Find streams with `aws logs describe-log-streams --log-group-name /aws/sagemaker/ProcessingJobs --log-stream-name-prefix <name> --region <region>`.
4. Read only matching streams with `aws logs filter-log-events --log-group-name /aws/sagemaker/ProcessingJobs --log-stream-names <stream...> --region <region>`.
5. For a stuck job, compare creation/start/end timestamps and inspect compute metrics for the exact window.
Do not stop/relaunch the job, edit IAM, or change query/container configuration without explicit approval.
## Classification
| Signature/evidence | Category | Likely fix |
|---|---|---|
| `ResourceLimitExceeded` or quota below requested count | Quota | Request quota or choose approved compute |
| `AccessDenied`, `iam:PassRole`, S3/ECR/KMS/Athena/Redshift denial | IAM | Correct role trust and add only missing resource-scoped permissions |
| Missing S3 object, invalid `/opt/ml/processing/` path, schema/format error | Data | Correct input/output mapping and script expectations |
| exit 137, OOM, disk-full, memory/disk saturation | Resources | Increase verified memory/volume or reduce working set/partitions |
| Image pull, entrypoint, module, dependency, nonzero exit | Container/user code | Verify image, command, script, source directory, and requirements |
| Athena/Redshift query syntax, workgroup, cluster, or result-location failure | Dataset definition | Validate SQL, metadata, result S3 URI, and warehouse permissions |
| Spark executor lost, serialization, shuffle/disk failure | Spark | Validate dependency packaging, partitions, executor memory, and instance count |
| Long `InProgress` with no logs/metrics movement | Stuck | Check process liveness and data volume; stop only after user approval |
## Response format
Return:
- job name/ARN, status, and timeline
- failure signature and root-cause category
- evidence from `ExitMessage`/`FailureReason`, decisive logs, and metrics if checked
- concrete repair steps ordered by lowest risk
- an explicit **Not checked** section
If evidence is insufficient, state the next read-only check instead of guessing.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |