New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Warehouse SQL

by HybridAIOne in HybridAIOne/hybridclaw

Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants.

MITAuto-check passedDatabases

Install Warehouse SQL

skills CLI
$ npx skills add HybridAIOne/hybridclaw --skill warehouse-sql -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install HybridAIOne/hybridclaw warehouse-sql --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/HybridAIOne/hybridclaw.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/warehouse-sql .claude/skills/warehouse-sql && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
warehouse-sql
GitHub stars
158
Token cost
~1.8k tokens
SKILL.md length
688 words
Files
6 (incl. scripts, references)
Skills in repo
70
Repo updated
First seen
Licence
MIT

At a glance

Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants.

  • Works in 4 steps: Refresh or read cached schema before… → Have the model draft SQL using the… → Return the SQL to the user before… → …
  • Tasks that involve SQL
  • SKILL.md covers Scope, Default Workflow, Backend Contract and Schema Cache, plus 4 more sections
  • Runs Python scripts from its folder; calls python3; needs HYBRIDCLAW_GATEWAY_TOKEN and GATEWAY_API_TOKEN

What it does

Warehouse SQL is an agent skill from HybridAIOne/hybridclaw. Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including scripts and reference files (for example `evals/tpch_eval_planner.py`, `evals/tpch_scenarios.json` and `references/backend-contract.md`).

It sits in Databases, covering SQL and Data warehousing. It works with SQL, ClickHouse, Google BigQuery and Snowflake. The repository describes itself as: Enterprise-ready self-hosted AI assistant runtime with sandboxed execution, secure credentials, approvals, and memory. The licence is MIT.

When your agent uses it

  • Tasks that involve SQL
  • Tasks that involve Data warehousing

Example prompts

  • “/warehouse-sql”

Requirements

  • Python 3
  • A credential in HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN
  • A credential in HYBRIDCLAW_GATEWAY_TOKEN

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Refresh or read cached schema before asking the model to draft SQL
  2. Have the model draft SQL using the cached schema, then review it before execution
  3. Return the SQL to the user before execution when the user asks for review,
  4. Execute only after the SQL review passes and include the original question

What it can do on your machine

Read from SKILL.md and the folder at commit 48ac731. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • HYBRIDCLAW_GATEWAY_TOKEN
    • GATEWAY_API_TOKEN
    • HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Warehouse SQL loads about 1.8k tokens when it runs, and up to ~2.3k if it reads all its reference files. Until then it costs about 38 tokens; SKILL.md has 688 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~38
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from HybridAIOne/hybridclaw at commit 48ac731, republished under its MIT licence (© HybridAIOne). 688 words, ~1,846 tokens.

Download SKILL.mdSave it as .claude/skills/warehouse-sql/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
warehouse-sql
description
Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants.
user-invocable
true
requires.bins
python3

Warehouse SQL

Use this skill when the user asks natural-language questions of a data warehouse, analytics database, or TPC-H-style reporting dataset.

Scope

  • schema introspection for SQLite eval databases and pluggable Postgres, ClickHouse, BigQuery, and Snowflake backends
  • cached schema summaries with explicit refresh commands for scheduled runs
  • reproducible TPC-H-style evaluation cases for generated SQL
  • deterministic SQL safety review before execution
  • read-only execution by default
  • write detection and explicit per-skill grant checks before any mutation

Default Workflow

  1. Refresh or read cached schema before asking the model to draft SQL:
    bash
    python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json schema --backend sqlite --database ./warehouse.db
  2. Have the model draft SQL using the cached schema, then review it before execution:
    bash
    python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json review "SELECT c_name FROM customer LIMIT 10"
    For review-only commands, pass --model-review with the original question to invoke HybridClaw's OpenAI-compatible gateway for the business-meaning review:
    bash
    python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json review --model-review --question "Show the first 10 customers" "SELECT c_name FROM customer LIMIT 10"
  3. Return the SQL to the user before execution when the user asks for review, when the query is broad, or when the result could expose sensitive business data.
  4. Execute only after the SQL review passes and include the original question:
    bash
    python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json query --backend sqlite --database ./warehouse.db --execute --question "Show the first 10 customers" "SELECT c_name FROM customer LIMIT 10"
    query --execute always invokes model review before running SQL. Configure model review with HYBRIDCLAW_GATEWAY_URL / GATEWAY_BASE_URL and HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN, HYBRIDCLAW_GATEWAY_TOKEN, or GATEWAY_API_TOKEN. Execution requires --question so the model can check whether the SQL answers the user's request.

Backend Contract

Supported backend names:

  • sqlite — executable through Python stdlib; used by the bundled eval suite
  • postgres — psycopg driver, or a connector command
  • clickhouse — clickhouse-connect driver, or a connector command
  • bigquery — google-cloud-bigquery driver, or a connector command
  • snowflake — snowflake-connector-python driver, or a connector command

For non-SQLite warehouses, install the backend driver package in the skill runtime or provide --backend-command / HYBRIDCLAW_WAREHOUSE_SQL_<BACKEND>_COMMAND. Connector commands read SQL on stdin and emit a JSON row array, {"rows": [...]} or CSV with headers. Do not invent credentials.

BigQuery schema introspection requires --bigquery-dataset or HYBRIDCLAW_WAREHOUSE_SQL_BIGQUERY_DATASET. Set --bigquery-project / HYBRIDCLAW_WAREHOUSE_SQL_BIGQUERY_PROJECT when the dataset is not in the default project.

Schema Cache

The helper writes schema cache files outside the repo by default at:

text
~/.hybridclaw/warehouse-sql/schema-cache

Use --cache-dir for tests or customer-specific workspaces. Use --refresh to force re-introspection. Use schedule-refresh to register the refresh with the HybridClaw gateway scheduler:

bash
python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json schedule-refresh --backend postgres --profile analytics --every "0 */6 * * *"

Scheduled refreshes default to last-channel delivery. Use --delivery-kind channel --delivery-to <channel-id> only when the target channel id is known.

Set HYBRIDCLAW_GATEWAY_TOKEN or GATEWAY_API_TOKEN in the environment for production scheduler registration. --gateway-token is supported for tests, but tokens passed as CLI arguments can be visible in process listings.

Show full SKILL.md (306 more words)Show less

Read/Write Rules

  • Default posture is read-only. SELECT, WITH, and EXPLAIN are allowed.
  • Mutating SQL is blocked unless all of these are true:
    • the user explicitly requested a write
    • the command includes --allow-write
    • the environment contains HYBRIDCLAW_WAREHOUSE_SQL_WRITE_GRANT
    • --write-grant exactly matches that environment value
  • Do not set or reveal the write grant in chat. Treat it like an operator capability, not a user-facing token.
  • Even with a write grant, explain the mutation and ask for confirmation before running it unless the user already gave a concrete write instruction.

SQL Review Rules

Before execution, check:

  • single statement only
  • read-only unless the explicit write grant path is used
  • selected columns are narrow enough for the task
  • LIMIT is present for exploratory row reads
  • joins use known keys from the schema cache
  • date and tenant filters are present when the question implies scope

The helper emits a review object with safety status and findings. Treat that as a deterministic guardrail. Pass --model-review on review-only commands to have the helper invoke the configured OpenAI-compatible model endpoint and add that verdict to review.modelReview; query --execute invokes model review automatically. Execution is blocked unless both the deterministic guardrail and model review pass.

Model review defaults to HYBRIDCLAW_GATEWAY_URL / GATEWAY_BASE_URL plus /v1/chat/completions and model auxiliary/eval_judge, with authentication from HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN, HYBRIDCLAW_GATEWAY_TOKEN, or GATEWAY_API_TOKEN. Use --model-review-url, --model-review-model, --schema-cache, and --question to make the review explicit in tests or operator workflows.

Eval Suite

Run the offline TPC-H-style scenarios:

bash
python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json eval-scenarios

The fixture at evals/tpch_tiny.sql contains a tiny public-schema-compatible dataset using TPC-H-style tables (customer, orders, lineitem, supplier, part, nation). It is an offline deterministic fixture for SQL generation coverage, not a TPC-H benchmark run. The scenario file at evals/tpch_scenarios.json verifies read-only review and execution against deterministic answers.

To measure model SQL generation against the same deterministic fixture, run the model-backed planner:

bash
python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json eval-scenarios --model-planner --model-review-model gpt-5

Use --scenario-id <id> to isolate one scenario while iterating.

Validation

Run:

bash
python3 skills/skill-creator/scripts/quick_validate.py skills/warehouse-sql
python3 skills/warehouse-sql/scripts/warehouse_sql.py --help
python3 skills/warehouse-sql/scripts/warehouse_sql.py --format json eval-scenarios

© HybridAIOne, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 5 other files (scripts, references) in skills/warehouse-sql of HybridAIOne/hybridclaw.

  • SKILL.md
  • evals/tpch_eval_planner.py
  • evals/tpch_scenarios.json
  • evals/tpch_tiny.sql
  • references/backend-contract.md
  • scripts/warehouse_sql.py

Open the folder on GitHubat commit 48ac731

Compare with similar skills

Warehouse SQL next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Warehouse SQL compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Warehouse SQL this skillHybridAIOne/hybridclaw158—~1.8kAutomated safety check: PassMIT
Semantic Analystsidequery/sidemantic129—~982Automated safety check: PassAGPL-3.0
Database MigrationRain-kl/OpenFlare288—~1.3kAutomated safety check: PassApache-2.0
SQL Queriesw95/awesome-claude-corporate-skills2373 repos~2.8kAutomated safety check: PassMIT
SQL Sentinelsickn33/agentic-awesome-skills47k1 repos~1.5kAutomated safety check: PassMIT
SQL Query Explainermohitagw15856/pm-claude-skills1.4k—~1.6kAutomated safety check: PassMIT

Similar skills

  • Semantic Analyst

    sidequery/sidemantic

    Answer analytical, KPI, metric, trend, cohort, and business-performance questions through a Sidemantic semantic layer.

    129 GitHub stars~982 tokensUpdated yesterday
    DatabasesAuto-check passed

  • Database Migration

    Rain-kl/OpenFlare

    Wavelet 项目专用：当新增或修改数据库表结构、索引、初始化数据、系统配置 seed、模板 seed、默认管理员、goose SQL 迁移、internal/infra/persistence/migrator、ClickHouse 分析库 DDL 或数据库升级流程时必须使用。本技能指导在 internal/infra/persistence/migrator/goose 下编写…

    288 GitHub stars~1.3k tokensUpdated yesterday
    DatabasesAuto-check passed

  • SQL Queries

    w95/awesome-claude-corporate-skills

    Write correct, performant SQL across all major data warehouse dialects (Snowflake, BigQuery, Databricks, PostgreSQL, etc.).

    237 GitHub starsUsed in 3 repos~2.8k tokens
    DatabasesAuto-check passed

  • SQL Sentinel

    sickn33/agentic-awesome-skills

    Audit SQL for the cost & performance anti-patterns that burn warehouse credits.

    47k GitHub starsUsed in 1 repo~1.5k tokens
    DatabasesAuto-check passed

  • SQL Query Explainer

    mohitagw15856/pm-claude-skills

    Explains, optimises, writes, and documents SQL queries. An agent skill from mohitagw15856/pm-claude-skills.

    1.4k GitHub stars~1.6k tokensUpdated yesterday
    DatabasesAuto-check passed

  • dbt Snowflake to BigQuery Translator

    google/skills

    Official

    Translates Snowflake dbt SQL models into standardized BigQuery SQL, keeping Jinja constructs and tracking progress in a migration tasks file.

    21k GitHub stars~2.7k tokensUpdated yesterday
    DatabasesAuto-check passed

More from HybridAIOne/hybridclaw

All 70 skills in this repo

  • Hermes3000 Writing

    HybridAIOne/hybridclaw

    Use Hermes3000 to plan, draft, revise, save, check consistency, and export long-form manuscripts through the Hermes3000 AI writing portal API.

    158 GitHub stars~2.7k tokensUpdated yesterday
    Auto-check passed

  • Manim Video

    HybridAIOne/hybridclaw

    Plan, script, render, and stitch Manim Community Edition videos in Python.

    158 GitHub stars~4.7k tokensUpdated yesterday
    Auto-check: notes

  • Skill Creator

    HybridAIOne/hybridclaw

    Create and update SKILL.md-based skills with strong trigger metadata, lean docs, and reliable init, validate, package, and publish workflows.

    158 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed

  • XLSX

    HybridAIOne/hybridclaw

    Create, edit, inspect, and analyze .xlsx spreadsheets and Excel workbooks.

    158 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed

  • Excalidraw

    HybridAIOne/hybridclaw

    Create and revise editable .excalidraw diagrams as Excalidraw JSON for architecture diagrams, flowcharts, sequence diagrams, concept maps, and other hand-drawn explainers.

    158 GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed

  • Google Ads

    HybridAIOne/hybridclaw

    Manage Google Ads accounts with safe GAQL reporting, campaign planning, guarded mutations, and gateway-proxied REST API calls.

    158 GitHub stars~4k tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Warehouse SQL

What does Warehouse SQL do?

Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants. Warehouse SQL is an agent skill from HybridAIOne/hybridclaw. Review and run read-only natural-language SQL against a customer data warehouse with cached schema introspection and explicit write grants.

When should I use Warehouse SQL?

Warehouse SQL fits situations like: tasks that involve SQL; tasks that involve Data warehousing.

How do I install Warehouse SQL in Claude Code?

Run `npx skills add HybridAIOne/hybridclaw --skill warehouse-sql -a claude-code`. Or copy the skill folder (skills/warehouse-sql in HybridAIOne/hybridclaw) into .claude/skills/warehouse-sql in your project. Claude Code loads it when a task matches its description.

How do I install Warehouse SQL in Codex?

Run `npx skills add HybridAIOne/hybridclaw --skill warehouse-sql -a codex`. Or copy the skill folder (skills/warehouse-sql in HybridAIOne/hybridclaw) into .agents/skills/warehouse-sql in your project. Codex loads it when a task matches its description.

Can I use Warehouse SQL in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add HybridAIOne/hybridclaw --skill warehouse-sql -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/warehouse-sql, .gemini/skills/warehouse-sql, .github/skills/warehouse-sql and .opencode/skills/warehouse-sql in your project.

What does Warehouse SQL need to run?

Going by SKILL.md and its folder, Warehouse SQL needs Python for the scripts in its folder, the command-line tools its instructions call (python3) and credentials named HYBRIDCLAW_GATEWAY_TOKEN, GATEWAY_API_TOKEN and HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN. Our summary lists: Python 3; A credential in HYBRIDCLAW_WAREHOUSE_SQL_MODEL_REVIEW_TOKEN; A credential in HYBRIDCLAW_GATEWAY_TOKEN.

Does Warehouse SQL access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Warehouse SQL safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Warehouse SQL use?

Warehouse SQL is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Warehouse SQL use?

About 1.8k tokens (SKILL.md is roughly 7.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 423 tokens, read only when the agent opens those files.

What are the alternatives to Warehouse SQL?

Skills that share tags, products or a category with Warehouse SQL: Semantic Analyst (sidequery/sidemantic, 129 stars), Database Migration (Rain-kl/OpenFlare, 288 stars), SQL Queries (w95/awesome-claude-corporate-skills, 237 stars) and SQL Sentinel (sickn33/agentic-awesome-skills, 47k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Warehouse SQL?

HybridAIOne (a GitHub organization) maintains it in HybridAIOne/hybridclaw, which has 158 GitHub stars. The repository holds 70 skills in this directory. The repository was last updated on October 7, 2026.

Source: HybridAIOne/hybridclaw on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.