PrismorSecPrismorSecOfficialยท2 Agent Skills Included

prismor

Runtime security guardrails and secret protection for coding agents

Blocks destructive commands, prompt injection, and secret leaks from coding agents in real time. Scores package installs against a live threat feed and keeps real credentials out of model context. Replaces manual security reviews with policy-driven enforcement, session audits, and a local dashboard.
npx skills add PrismorSec/prismor --all -g -y
Available:

Instructs the agent to preserve Prismor's security posture by keeping the advisory feed, skill instructions, and runtime policy engine aligned, and to never expose real secret values when editing the cloaking subsystem.

All Skills in This Repository (2)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Prismor?โ–ผ

Run `npx skills add PrismorSec/prismor --all -g -y` in your terminal to install all skills in this suite globally.

How to stop AI coding agents from leaking secrets?โ–ผ

Prismor's cloaking layer stores real credentials locally and swaps them in at execution time, so secrets never enter model context or transcripts.

Can Prismor block dangerous shell commands?โ–ผ

Yes. Its policy engine intercepts tool calls and blocks destructive commands, privilege escalation, and prompt injection before they run.

Which AI coding agents does Prismor support?โ–ผ

It works with Claude Code, Cursor, Codex, Windsurf, Copilot CLI, and over a dozen other agents through hook-based integrations.

Does Prismor check package installs for supply chain risks?โ–ผ

Yes. Wrapping installs with `prismor supplychain` scores packages against a live IOC feed and maintainer signals before anything is installed.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’