oxy-core

A set of conventions for Oxy data projects, a framework for semantic layers, SQL workflows, AI agents, and dashboards. It explains what the project's main file types and folders mean.

In plain words
What is it for?
Use it as a reference while planning or changing Oxy projects, including configuration, semantic-layer files, workflows, agents, dashboards, and ETL pipelines.
Why use it?
It helps developers choose the right Oxy feature for a task and keep files organized, especially when deciding between a semantic query, SQL, workflow, or agent.

Cursor rule

Install

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.

agentmods
npx agentmods add rules/oxy-hq/skills/oxy-core
Clone the repo
git clone --depth 1 https://github.com/oxy-hq/skills
Per session 989 This file is loaded in full into every session.
When invoked 989 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00989 $0.00989
Opus 5 $0.00495 $0.00495
Sonnet 5 $0.00198 $0.00198
Haiku 4.5 $0.00099 $0.00099

Measured 2d ago against content hash b62a57791571, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

oxy-core 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 2d 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.

cursor/rules/oxy-core.mdc · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Oxy Project Conventions

You are working in an Oxy data project. Oxy is a framework for building data apps with semantic layers, SQL workflows, AI agents, and interactive dashboards.

File Conventions

Pattern Purpose
config.yml Project configuration (databases, settings)
semantics.yml Semantic layer entry point
*.view.yml Semantic layer views (entities, dimensions, measures)
*.topic.yml Semantic layer topics (organize views by domain)
*.workflow.yml Multi-step data pipelines
*.agent.yml AI-powered analysis agents
*.app.yml Data app dashboards (tasks + displays)
.databases/ Auto-generated database schemas (from oxy sync)
etl/ ETL pipelines using DLT

The Oxy Hierarchy

When solving data analysis problems, follow this priority order:

Does the semantic layer have the data?
├─ YES -> Use semantic queries (PREFERRED)
└─ NO -> Is this a deterministic query/pipeline?
    ├─ YES -> Use SQL file or Workflow
    └─ NO -> Need AI reasoning?
        ├─ YES -> Use an Agent
        └─ NO -> Build semantic layer views first, then use semantic queries
  1. Semantic Queries (preferred): Use when views/topics exist covering the data. Natural language, automatic joins, centrally maintained.
  2. SQL / Workflows (fallback): Use when data isn't in semantic layer, or you need custom logic, ETL, or parameterized queries.
  3. AI Agents (last resort): Use only when you need AI reasoning, exploratory analysis, or dynamic query generation.

Always check if semantic layer views exist before writing SQL or agents.

CLI Commands Reference

Command Purpose
oxy sync Extract database schemas to .databases/
oxy sync <db-name> Sync a specific database
oxy validate Validate all YAML configs (agents, workflows, apps, semantic layer)
oxy validate --file=<path> Validate a single file
oxy build Compile semantic layer and build vector embeddings
oxy run query.sql Execute a SQL file
oxy run query.sql --dry-run Test SQL without executing
oxy run query.sql -v key=value Run SQL with variables
oxy run pipeline.workflow.yml Run a workflow
oxy run agent.agent.yml "question" Run an agent with a prompt
oxy test <file> Run evaluation tests on agent/workflow
oxy semantic-engine --dev-mode Start semantic engine for testing queries
oxy start Start the web UI for rendering apps

Read the full file on GitHub · 96 lines

Changes

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.

  1. 2d ago First seen · 96 lines · 989 tokens per session scan A b62a57791571

Subscribe to this mod's changes

oxy-core is a cursor rule published in the GitHub repository oxy-hq/skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 989 tokens to every session, about $0.0049 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.