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 skills add vasilyu1983/AI-Agents-public --skill data-lake-platformgit clone --depth 1 https://github.com/vasilyu1983/AI-Agents-publicWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vasilyu1983/ai-agents-public/data-lake-platform)<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/data-lake-platform"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/data-lake-platform/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vasilyu1983/ai-agents-public/data-lake-platform"><img src="https://agentmods.dev/badge/skills/vasilyu1983/ai-agents-public/data-lake-platform.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 48 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 194 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 208 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What 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 5.1 | $0.00041 | $0.04548 |
| Opus 5.5 | $0.00016 | $0.01819 |
| Sonnet 5.5 | $0.00008 | $0.00910 |
| Haiku 4.5 | $0.00004 | $0.00455 |
Grade A, and why
data-lake-platform 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 22d ago.
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 — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Data Lake Platform
Build and operate production data lakes and lakehouses: ingest, store in open table formats, catalog, maintain, and serve analytics reliably. Transformation (dbt/SQLMesh) and data-quality testing live in data-analytics-engineering; Kafka/Flink streaming lives in data-streaming.
Quick Reference
| Task | Resource |
|---|---|
| Pick a table format and spec version | references/storage-formats.md |
| Pick a catalog / control plane | references/governance-catalog.md |
| Design a batch or CDC ingest path | references/ingestion-patterns.md |
| Scaffold or inspect an Iceberg table | scripts/scaffold_iceberg_table.py, scripts/inspect_iceberg_metadata.sh |
| Check any version, spec-support, or vendor feature-status claim | Version and support lookup |
Quick-Start Decision Table
| Situation | Default choice |
|---|---|
| Open multi-engine analytics | Iceberg + REST catalog (Polaris/Glue/Nessie) + Trino/Spark |
| Databricks primary compute | Delta + Unity Catalog; add UniForm only if external readers exist |
| CDC-heavy mutable, Spark-centered | Hudi (CoW for read-heavy, MoR for write-heavy) + Kafka/Debezium |
| Streaming-first mutable, Flink-centered | Paimon + Flink CDC |
| Single-engine embedded / analyst workstation | DuckLake, or DuckDB + Parquet/Iceberg |
| Low-latency BI, high-concurrency dashboards | ClickHouse or StarRocks serving layer, fed from the lake |
| Heterogeneous estate (Hive + Kafka + RDBMS) | Apache Gravitino as a federated "catalog of catalogs" |
Format Comparison
| Property | Apache Iceberg | Delta Lake | Apache Hudi | Apache Paimon |
|---|---|---|---|---|
| Spec / protocol | Format v2 is the broad-compatibility baseline; v3 support is per engine (look up) | Protocol features gated per table (reader/writer features) | Table version per release line | Flink-first LSM table format |
| Row-level deletes | Position/equality delete files (v2); deletion vectors (v3 spec) | Deletion vectors | MoR log files | LSM merge |
| Catalog posture | REST-catalog-first | Catalog-managed tables (Unity Catalog) | Mixed | Flink catalog first |
| Branches and tags | First-class | Limited | Workflow-specific | Workflow-specific |
| Multi-engine reads | Strongest (Trino, Spark, DuckDB, Flink) | Via UniForm and native connectors | Narrower; validate | Narrower; validate |
What ships with it
54 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 338 B
- assets/cross-platform/template-cost-optimization.md 5.3 KB
- assets/cross-platform/template-data-pipeline.md 4.1 KB
- assets/cross-platform/template-data-quality-backfill-runbook.md 3.0 KB
- assets/cross-platform/template-data-quality-governance.md 3.5 KB
- assets/cross-platform/template-data-quality.md 1.0 KB
- assets/cross-platform/template-incremental-loading.md 6.3 KB
- assets/cross-platform/template-ingestion-governance-checklist.md 3.0 KB
- assets/cross-platform/template-medallion-architecture.md 2.5 KB
- assets/cross-platform/template-migration-checklist.md 3.1 KB
- assets/cross-platform/template-partitioning-strategy.md 5.1 KB
- assets/cross-platform/template-schema-evolution.md 5.5 KB
- assets/ingestion/airbyte/template-airbyte-connection.md 4.8 KB
- assets/ingestion/dlt/template-dlt-database-source.md 4.5 KB
- assets/ingestion/dlt/template-dlt-incremental.md 5.7 KB
- assets/ingestion/dlt/template-dlt-pipeline.md 5.3 KB
- assets/ingestion/dlt/template-dlt-rest-api.md 4.6 KB
- assets/ingestion/dlt/template-dlt-warehouse-loading.md 4.5 KB
- assets/orchestration/template-dagster-pipeline.md 4.7 KB
- assets/query-engines/clickhouse/template-clickhouse-ingestion.md 5.5 KB
- assets/query-engines/clickhouse/template-clickhouse-materialized-views.md 6.0 KB
- assets/query-engines/clickhouse/template-clickhouse-optimization.md 7.0 KB
- assets/query-engines/clickhouse/template-clickhouse-replication.md 5.9 KB
- assets/query-engines/clickhouse/template-clickhouse-setup.md 5.0 KB
- assets/query-engines/doris/template-doris-setup.md 4.7 KB
- assets/query-engines/duckdb/template-duckdb-analytics.md 4.3 KB
- assets/query-engines/duckdb/template-duckdb-data-import.md 5.3 KB
- assets/query-engines/starrocks/template-starrocks-setup.md 4.2 KB
- assets/storage/delta/template-delta-table.md 5.8 KB
- assets/storage/hudi/template-hudi-table.md 6.1 KB
- assets/storage/iceberg/template-iceberg-maintenance.md 7.7 KB
- assets/storage/iceberg/template-iceberg-table.md 7.5 KB
- assets/storage/parquet/template-parquet-optimization.md 5.6 KB
- assets/visualization/metabase/connection-checklist.md 780 B
- assets/visualization/metabase/dashboard-request.md 632 B
- assets/visualization/metabase/incident-playbook.md 1.0 KB
- data/sources.json 18 KB
- learnings.consolidated.md 594 B
- learnings.md 465 B
- references/architecture-patterns.md 3.2 KB
- references/bi-visualization-patterns.md 2.4 KB
- references/cost-optimization.md 2.7 KB
- references/data-mesh-patterns.md 4.7 KB
- references/governance-catalog.md 4.2 KB
- references/ingestion-patterns.md 4.3 KB
- references/operational-playbook.md 4.0 KB
- references/orchestration-patterns.md 2.6 KB
- references/query-engine-patterns.md 3.1 KB
- references/security-access-patterns.md 4.7 KB
- references/storage-formats.md 7.6 KB
- scripts/inspect_iceberg_metadata.sh 9.4 KB runs code
- scripts/scaffold_iceberg_table.py 17 KB runs code
- scripts/test_inspect_iceberg_metadata.py 3.2 KB runs code
- scripts/test_scaffold_iceberg_table.py 1.0 KB runs code
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.
- 22d ago Changed · -41 lines 2b10d143cbac
- 1mo ago Changed · +11 lines ca93fe122718
- 1mo ago First seen · 269 lines · 41 tokens per session scan A 18b96db65785
data-lake-platform is a skill published in the GitHub repository vasilyu1983/AI-Agents-public (101 stars, last pushed 11d ago), licensed MIT. It adds 41 tokens to every session and 4,548 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
AgentDB Performance Optimization
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.
AgentDB Vector Search
Implement semantic vector search with AgentDB for intelligent document retrieval, similarity matching, and context-aware querying. Use when building RAG systems, semantic search engines, or intelligent knowledge bases.
qdrant
Vector search engine for production RAG systems.
chroma
Embedding database for RAG and semantic search.
pinecone
Managed vector DB for production RAG and search.
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.