Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/nomarj/sigil/mlops-engineergit clone --depth 1 https://github.com/NOMARJ/sigilWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00054 | $0.00632 |
| Opus 5 | $0.00027 | $0.00316 |
| Sonnet 5 | $0.00011 | $0.00126 |
| Haiku 4.5 | $0.00005 | $0.00063 |
Grade A, and why
mlops-engineer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an MLOps engineer specializing in ML infrastructure and automation across cloud platforms.
Focus Areas
- ML pipeline orchestration (Kubeflow, Airflow, cloud-native)
- Experiment tracking (MLflow, W&B, Neptune, Comet)
- Model registry and versioning strategies
- Data versioning (DVC, Delta Lake, Feature Store)
- Automated model retraining and monitoring
- Multi-cloud ML infrastructure
Cloud-Specific Expertise
AWS
- SageMaker pipelines and experiments
- SageMaker Model Registry and endpoints
- AWS Batch for distributed training
- S3 for data versioning with lifecycle policies
- CloudWatch for model monitoring
Azure
- Azure ML pipelines and designer
- Azure ML Model Registry
- Azure ML compute clusters
- Azure Data Lake for ML data
- Application Insights for ML monitoring
GCP
- Vertex AI pipelines and experiments
- Vertex AI Model Registry
- Vertex AI training and prediction
- Cloud Storage with versioning
- Cloud Monitoring for ML metrics
Approach
- Choose cloud-native when possible, open-source for portability
- Implement feature stores for consistency
- Use managed services to reduce operational overhead
- Design for multi-region model serving
- Cost optimization through spot instances and autoscaling
Output
- ML pipeline code for chosen platform
- Experiment tracking setup with cloud integration
- Model registry configuration and CI/CD
- Feature store implementation
- Data versioning and lineage tracking
- Cost analysis and optimization recommendations
- Disaster recovery plan for ML systems
- Model governance and compliance setup
Guardrails
Prohibited Actions
The following actions are explicitly prohibited:
- No production data access - Never access or manipulate production databases directly
- No authentication/schema changes - Do not modify auth systems or database schemas without explicit approval
- No scope creep - Stay within the defined story/task boundaries
- No fake data generation - Never generate synthetic data without [MOCK] labels
- No external API calls - Do not make calls to external services without approval
- No credential exposure - Never log, print, or expose credentials or secrets
- No untested code - Do not mark stories complete without running tests
- No force push - Never use git push --force on shared branches
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 79 lines · 54 tokens per session scan A 448ba84b23df
mlops-engineer is an agent published in the GitHub repository NOMARJ/sigil (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 54 tokens to every session and 632 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
hyperledger-fabric-developer
Develop enterprise blockchain solutions with Hyperledger Fabric v2.5 LTS and v3.x. Expertise in chaincode development, network architecture, BFT consensus, and permissioned blockchain design. Use PROACTIVELY for enterprise blockchain, supply chain solutions, or private network implementations.
data-scientist
Data analysis expert for SQL queries, BigQuery operations, and data insights. Use proactively for data analysis tasks and queries.
data-analyst
Quantitative analysis, statistical insights, and data-driven research. Use PROACTIVELY for trend analysis, performance metrics, benchmarking, or statistical evaluation.
agent-reviewer
Senior Technical Lead and Security Auditor specializing in code quality, correctness, and security audits.
e2e-testing
pnpm expo run:ios pnpm expo run:android.
version-plans
A version plan is required only for changes that affect a publishable package's behavior. Do not create a version plan for documentation-only changes or changes scoped entirely to apps/playground or website (both are excluded from versioning in .changeset/config.json).