accidental-data-loss-prevention

Blocks irreversible deletion commands until explicit user consent is obtained.

Updated May 26, 2026
One-click install
npx skills add https://github.com/KinalVR-Expo2026/KinalVR-Client --skill accidental-data-loss-prevention-kinalvr-expo2026
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: accidental-data-loss-prevention
Source: https://github.com/KinalVR-Expo2026/KinalVR-Client/tree/main/.claude/.agents/skills/accidental-data-loss-prevention
Command: npx skills add https://github.com/KinalVR-Expo2026/KinalVR-Client --skill accidental-data-loss-prevention-kinalvr-expo2026

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents can execute destructive commands like DROP TABLE, TRUNCATE, or gsutil rm without pausing, causing irreversible loss of production data or critical infrastructure. This Skill enforces a mandatory stop-and-verify checkpoint before any such command runs. ## Core Features & Use Cases - Mandatory Consent Gate: Halts execution of any irreversible deletion command and requires explicit user approval before proceeding. - Broad Coverage of Destructive Operations: Covers SQL (DROP TABLE/VIEW/SCHEMA/DATABASE, TRUNCATE, unscoped DELETE), cloud storage deletion (gsutil rm, gcloud storage rm), and infrastructure teardown (gcloud projects delete, Spanner/BigQuery/Dataproc deletion, secret and KMS key destruction). - Use Case: When an agent is about to run gcloud projects delete my-prod-project during cleanup, it instead stops, explains the impact, and waits for your explicit confirmation. ## Quick Start Before running any destructive SQL, cloud storage, or infrastructure deletion command, stop and ask me for explicit confirmation first.

Frequently Asked Questions about accidental-data-loss-prevention

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent an AI agent from deleting production data?

Use a mandatory consent checkpoint that halts any irreversible deletion command before execution. The agent must explain the impact, state why deletion is necessary, and wait for your explicit approval before proceeding.

What SQL commands are considered destructive?

Destructive SQL includes DROP TABLE, DROP VIEW, DROP SCHEMA, DROP DATABASE, TRUNCATE, and broad DELETE statements missing a WHERE clause or using conditions like 1=1. These operations permanently remove data without a recovery path.

Does this cover Google Cloud resource deletion?

Yes, it covers gcloud projects delete, gsutil rm and gcloud storage rm against critical buckets, and deletion of Spanner, BigQuery, and Dataproc resources. It also includes secret deletion and KMS key destruction.

When should an agent ask for confirmation before deleting?

An agent should ask whenever a command causes irreversible data loss or when there is any doubt about the target. Waiting for confirmation is always preferred over accidentally deleting production data or critical project assets.

What are the limitations of a consent-based deletion safeguard?

It relies on the agent correctly recognizing a command as destructive before running it, so novel or obfuscated deletion patterns may slip through. It also adds a manual approval step, which slows fully automated cleanup workflows.