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 ScalefreeCOM/datavault4dbt-agent-skills --skill using-datavault4dbtgit clone --depth 1 https://github.com/ScalefreeCOM/datavault4dbt-agent-skillsWrote 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/scalefreecom/datavault4dbt-agent-skills/using-datavault4dbt)<a href="https://agentmods.dev/skills/scalefreecom/datavault4dbt-agent-skills/using-datavault4dbt"><img src="https://agentmods.dev/badge/skills/scalefreecom/datavault4dbt-agent-skills/using-datavault4dbt/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/scalefreecom/datavault4dbt-agent-skills/using-datavault4dbt"><img src="https://agentmods.dev/badge/skills/scalefreecom/datavault4dbt-agent-skills/using-datavault4dbt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00102 | $0.01713 |
| Opus 5 | $0.00051 | $0.00856 |
| Sonnet 5 | $0.00020 | $0.00343 |
| Haiku 4.5 | $0.00010 | $0.00171 |
Grade A, and why
using-datavault4dbt 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 9d 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Using datavault4dbt
datavault4dbt is a dbt package that generates Data
Vault 2.0 SQL from compact YAML metadata. You write a {%- set yaml_metadata -%} block and call a
macro; the package generates the insert-only loading logic, hashing, and high-water-mark optimization.
Core principle: every model is staging-driven. The stage macro computes the hashkeys and
hashdiffs once; hubs, links, and satellites then just reference those pre-computed columns. Get staging
right and the raw vault falls into place.
When to use
- Building or editing datavault4dbt models: staging, hubs, links, satellites, PITs, snapshot control.
- Setting up a raw vault from flat source tables.
- Deciding which Data Vault entity type fits a source (hub vs link vs which satellite).
- Configuring the package:
packages.yml, global variables, hashing, per-adapter setup.
Do NOT use for general dbt modeling unrelated to Data Vault, or semantic-layer/metrics work.
The non-negotiable workflow
- Verify, don't guess. Macro parameter names and defaults must come from the installed package
(
dbt_packages/datavault4dbt/docsand.../macros), not memory. When unsure, read the macro. - One stage per source table. Staging is for hashing and light shaping (derived columns, prejoins, missing columns) — never for harmonizing or business logic.
- Build in layers, bottom-up: sources → staging (view) → raw vault hubs/links/satellites (incremental) → business vault PIT/dims/facts (table/view).
- Validate with the warehouse. After building, run
dbt build --select <model>+and confirm the hashkey uniqueness/not-null tests pass. Look at the data; don't assume.
Reference guides
Read the relevant guide when working on that part of the vault:
| Guide | Use when |
|---|---|
| references/project-layout.md | Setting up a new datavault4dbt project, folder structure, layer materializations |
| references/conventions-and-config.md | packages.yml, global variables, hash settings, naming conventions, per-adapter notes |
| references/staging.md | Writing a stage model — hashed_columns, derived_columns, prejoins, missing_columns, ghost records, multi-active |
| references/choosing-the-right-entity.md | Deciding hub vs link vs which satellite for a given source |
| references/hubs-and-links.md | hub, link, nh_link — including multi-source loading and rsrc_static |
| references/satellites.md | sat_v0/sat_v1 and the multi-active, effectivity, record-tracking, and non-historized variants |
| references/business-vault.md | PIT tables, snapshot control, the PIT cleanup hook |
What ships with it
7 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.
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.
- 9d ago First seen · 123 lines · 102 tokens per session scan A 7a3bab15555f
using-datavault4dbt is a skill published in the GitHub repository ScalefreeCOM/datavault4dbt-agent-skills (22 stars, last pushed 11d ago), licensed Apache-2.0. It adds 102 tokens to every session and 1,713 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
exasol-distributed-ml
Distributed machine learning, data mining, and iterative HPC with Exasol. Covers end-to-end ML pipelines (DISTRIBUTE BY + SET scripts + BucketFS), per-entity federated training with partialfit and ctx.reset(), batch inference, map-reduce ensemble training, distributed ensemble and SON algorithm for frequent itemset…
exasol-import
Use Exasol IMPORT and IMPORT INTO SQL plus exapump upload local file workflows for moving data into Exasol. Covers CSV, FBV, and Parquet, CREATE CONNECTION connection objects for import, cloud credential patterns for S3, Azure Blob Storage, and Google Cloud Storage (GCS), reject handling, and staging-based import…
exasol-notebook-connector-config
Set up notebook-connector configuration for Exasol AI workflows through the Secrets Python API. Covers secure config store (scs) creation, the AILabConfig and StorageBackend key sources, backend-specific config values such as dbhostname, dbschema, storagebackend, and huggingfacetoken, Python validation, and handoff to…
exasol-document-virtual-schemas
Use Exasol document-file virtual schemas for federated read-only access to object and file storage — S3, Google Cloud Storage, Azure Blob Storage, and Azure Data Lake Storage Gen2 document files. Covers document-file adapter family selection, the EDML MAPPING definition, CREATE VIRTUAL SCHEMA usage, refresh workflows…
exasol-transformers
Deploy and use the Exasol Transformers Extension (TE) for NLP inference inside Exasol with notebook-connector. Covers initializeteextension, deployscripts, PYTHON3TE language-container activation SQL, Hugging Face model-installation workflows, and the current TE SQL UDF surface.
nexus-data-engineer
Data engineering agent specializing in ETL/ELT pipelines, data quality, warehousing, and streaming. Use when you need to design or review data pipelines, ensure data integrity, detect schema drift, or evaluate tools like dbt, Airflow, and streaming platforms for EdTech and enterprise workloads.