PrismorSec/immunity-agent — GitHub trending stats & insights

Reach 125K+ monthly visitors

PrismorSec/immunity-agent

Self improving security layer for AI coding agents : Safe package recommendations, Secret protection, MCP/tools guardrails, Skill governance, runtime policy enforcement and full audit visibility with a self-serve dashboard.

Visit GitHub

Python

213

4 contributors

Apache License 2.0

website

Social mentions

Recent discussions about this repository across the web

This in itself has 10+ critical vulnerabilities for a simple calculator webapp

@prismor_dev · x.com

Repo deleted because it had all commits from claude Fable 5

@prismor_dev · x.com

Not if you redact your secrets with immuity-agent

@prismor_dev · x.com

hey yash we removed the "dangerous" from claude --model claude-fable-5 --dangerously-skip-permissions, would be great to have you try and share feedback in our open source project, you can see your…

@prismor_dev · x.com

we removed the "dangerous" from claude --model claude-fable-5 --dangerously-skip-permissions, would be great to have feedback

@prismor_dev · x.com

99% of vibe coders right now

@prismor_dev · x.com

Nice try lin! Blocked by:

@prismor_dev · x.com

Traditional IAM: you prove who you are. Agent IAM: you prove who you are, what you're doing, why you're doing it, who asked you to, and whether the chain of delegation from human to orchestrator to…

@prismor_dev · x.com

Open sourced version which self evolves in detecting these along with secret management

@prismor_dev · x.com

We built immunity-agent to secure coding agents, but beyond secret redaction and supply chain we also wanted to test how well it holds against prompt injection So we benchmarked it against 800+…

@prismor_dev · x.com

Social mentions

Repository activities