tophant-aitophant-aiOfficialยท4 Agent Skills Included

ClawVault

Shield AI agents from data leaks, injection, and dangerous commands

Protects AI agents by scanning prompts and traffic for API keys, PII, prompt injection, and dangerous commands before damage happens. Masks sensitive document content, blocks risky tool calls, and enforces token budgets through a local proxy and dashboard. Sends real-time security alerts and daily reports so you always know what your AI touched.
npx skills add tophant-ai/ClawVault --all -g -y

All Skills in This Repository (4)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install ClawVault?โ–ผ

Run `npx skills add tophant-ai/ClawVault --all -g -y` in your terminal to install all skills in this suite globally.

How to stop AI agents from leaking API keys?โ–ผ

ClawVault scans every prompt and response for credentials, PII, and secrets, then blocks or masks them before they reach the model.

Can ClawVault block prompt injection attacks?โ–ผ

Yes. Its detection engine catches role hijacking, instruction overrides, and data exfiltration attempts, and can block or sanitize them automatically.

Does ClawVault work with OpenClaw and Claude Code?โ–ผ

Yes. It integrates with OpenClaw sessions and plugins, and adds prompt and tool hooks to Claude Code for full protection.

Can I mask company documents before AI analysis?โ–ผ

Yes. The mask-project skill replaces company names, amounts, and person names with placeholders so originals never reach the model.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’