- Home
- Skills
- Data & Databases
- Data Contract Enforcer
Works with the AI Agents you already use
Data Contract Enforcer
Detect schema and semantic drift, write the contract, add the CI check that keeps it true.
$15
Data Contract Enforcer
Example session with this skill installed
Our orders pipeline sends events to a warehouse. Last week a dashboard showed revenue 100x too high for two days and nobody noticed. Here is the producer schema and the consumer SQL.
- Read your context and instructions
- Compiled the data contract enforcer
Worked example — illustrative data.
Data contract: orders.v1
Drift found
| Field | Declared | Observed | Consumer assumption | Risk |
|---|---|---|---|---|
| amount | integer | integer | pounds | high: producer emitted minor units after a refactor |
| status | 3 values | 5 values | 3 branches | medium: two new values fall through to default |
| order_id | unique | duplicated 0.4% | join key | high: sums inflated by 100x on some days |
Contract
dataset orders.v1, owner orders-team, freshness 15m, completeness 99.5%.
Fields: order_id (string, unique), status (enum of 5), amount_minor (integer, GBP minor units), created_at (timestamp, UTC).
Guarantees: one row per order_id per partition, no deletion of paid orders. Breaking changes listed explicitly.
Enforcement
Producer validates before write and quarantines invalid rows instead of dropping. CI contract test fails on field removal, type change or enum removal. Consumers generate types from the contract. Runtime checks alert on freshness and volume, since an empty table passes every schema check.
Migration
New major version for amount_minor; dual-publish for 30 days; reconciliation count of rows over 1 pound; removal gate is zero reads for a full cycle.
Owners
Producer: orders team. Consumers: finance, warehouse, support. Review quarterly.
Known gaps
Refunds stream not covered yet.
Connects securely to your tools. The creator never sees your data.
About this skill
Data contract enforcement: detect schema drift with impact, write the contract, and block the break in CI before it reaches a dashboard.
Use this skill when:
- a dashboard broke after a schema change
- add a data contract
- producers and consumers disagree
- catch silent data drift
Who it's for: Data platform and analytics engineering teams.
Example request
The revenue dashboard broke again. Find the drift.
What the example returns
- Producer renamed amount to amount_cents
- 3 consumers affected, 1 silently
- Contract plus CI check to block it
In the package: five files — SKILL.md with the trigger and workflow, README.md, references/METHOD.md (the full method), references/CHECKLIST.md (the pass/fail gate) and examples/EXAMPLE.md (a worked run you can copy).
Works with Claude Code, Cursor, Codex CLI and any agent that reads SKILL.md. Plain markdown: no API key, no account, no network access.

Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Data & Databases.
Requires 30 days of data profiles and query logs for drift detection. Does not provide vendor-specific SDK code.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 days ago
Permissions
Works only from exports you provide. No network calls, no credentials.