accidental-data-loss-prevention

Requires explicit user consent before executing irreversible data deletion commands.

1|Updated Aug 30, 2026
One-click install
npx skills add https://github.com/FeexSystems/3WM-SONIK-LABS --skill accidental-data-loss-prevention-feexsystems
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: accidental-data-loss-prevention
Source: https://github.com/FeexSystems/3WM-SONIK-LABS/tree/main/.gemini/skills/accidental-data-loss-prevention
Command: npx skills add https://github.com/FeexSystems/3WM-SONIK-LABS --skill accidental-data-loss-prevention-feexsystems

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Destructive commands like DROP TABLE, TRUNCATE, or gsutil rm can permanently erase production databases, cloud storage, and infrastructure in seconds. This Skill prevents accidental data loss by forcing a mandatory verification step before any irreversible deletion runs. ## Core Features & Use Cases - Destructive SQL Guard: Intercepts DROP TABLE/VIEW/SCHEMA/DATABASE, TRUNCATE, and broad DELETE statements missing a WHERE clause or using 1=1. - Cloud Resource Protection: Flags gsutil rm, gcloud storage rm, project deletion, Spanner/BigQuery/Dataproc resource deletion, secret deletion, and KMS key destruction. - Mandatory Consent Workflow: Halts execution, explains the impact and rationale, and waits for explicit affirmative user approval before proceeding. - Use Case: When asked to clean up a database, the agent pauses before running TRUNCATE on a production table, explains the consequences, and only proceeds after you confirm. ## Quick Start Ask the agent to delete a table or remove cloud storage objects, and it will stop to request your explicit confirmation before executing anything irreversible.

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 accidental deletion of production data with AI agents?▼

Use a consent-gating skill that halts any destructive command before execution. The agent must explain the impact, justify why deletion is needed, and wait for your explicit approval before running operations like DROP TABLE or gsutil rm.

What SQL commands are flagged as destructive?▼

The skill flags DROP TABLE, DROP VIEW, DROP SCHEMA, DROP DATABASE, TRUNCATE, and broad DELETE statements that lack a WHERE clause or use conditions like 1=1. These operations are irreversible and require explicit user consent.

Does this cover Google Cloud storage and infrastructure deletion?▼

Yes, it covers gsutil rm and gcloud storage rm targeting production buckets, gcloud projects delete, deletion of Spanner, BigQuery, and Dataproc resources, secret deletion, and KMS key destruction.

What happens when a destructive command is detected?▼

Execution halts immediately. The agent explains the impact of the deletion, states why it believes the action is necessary, and requests explicit approval. It only proceeds after receiving clear affirmative consent in the conversation.

When should I ask for confirmation even if a command seems safe?▼

When in doubt, always ask. The skill's guiding principle is that waiting for confirmation is better than accidentally deleting production data or critical project assets, even for ambiguous cases.