DEV Community

Umang Kumar profile picture

Umang Kumar

Founder of Cirvix: open-source runtime authorization for AI agents. Capability is not authority.

Joined Joined on  Personal website https://cirvix.com twitter website
Codex CLI and Gemini CLI Safety Settings: A Side-by-Side Hardening Guide
umangcirvix profile

Codex CLI and Gemini CLI Safety Settings: A Side-by-Side Hardening Guide

#ai #security #cli #devops
Comments
5 min read

Want to connect with Umang Kumar?

Create an account to connect with Umang Kumar. You can also sign in below to proceed if you already have an account.

Create Account
Already have an account? Sign in
git push --force Guardrails for AI Agents: Server, Client, and Policy Layers That Actually Hold
umangcirvix profile

git push --force Guardrails for AI Agents: Server, Client, and Policy Layers That Actually Hold

#git #ai #security #devops
Comments
5 min read
Human Approval for AI Agent Tool Calls: What to Hold, Who Approves, and How the Agent Waits
umangcirvix profile

Human Approval for AI Agent Tool Calls: What to Hold, Who Approves, and How the Agent Waits

#ai #security #agents #devops
Comments
5 min read
Audit Logs for AI Coding Agents: What to Record, Where to Collect It, and What It Proves
umangcirvix profile

Audit Logs for AI Coding Agents: What to Record, Where to Collect It, and What It Proves

#ai #security #devops #observability
Comments
5 min read
Protecting .env Secrets From AI Agents: Six Layers, and How to Test Each One
umangcirvix profile

Protecting .env Secrets From AI Agents: Six Layers, and How to Test Each One

#security #ai #devops #webdev
Comments
5 min read
Prompt Injection in Coding Agents: Where It Gets In and What Actually Limits the Damage
umangcirvix profile

Prompt Injection in Coding Agents: Where It Gets In and What Actually Limits the Damage

#ai #security #llm #programming
1
Comments 4
5 min read
Cursor Agent Permissions Explained: What Runs Without Asking, and How to Tighten It
umangcirvix profile

Cursor Agent Permissions Explained: What Runs Without Asking, and How to Tighten It

#cursor #security #ai #productivity
Comments
4 min read
MCP Server Security Risks: 7 Threats to Model Before You Connect One
umangcirvix profile

MCP Server Security Risks: 7 Threats to Model Before You Connect One

#mcp #security #ai #llm
Comments 3
5 min read
How to Stop an AI Agent From Running rm -rf (and Why String Matching Isn't Enough)
umangcirvix profile

How to Stop an AI Agent From Running rm -rf (and Why String Matching Isn't Enough)

#ai #security #bash #devops
Comments
5 min read
Claude Code Security Best Practices: A Practical Hardening Guide
umangcirvix profile

Claude Code Security Best Practices: A Practical Hardening Guide

#claudecode #security #ai #devops
Comments
5 min read
Hash-chained audit logs for AI agent actions, in 40 lines
umangcirvix profile

Hash-chained audit logs for AI agent actions, in 40 lines

#ai #security #mcp #devops
Comments
3 min read
Securing MCP servers: a practical checklist for 2026
umangcirvix profile

Securing MCP servers: a practical checklist for 2026

#ai #security #mcp #devops
2
Comments 6
4 min read
Why approval prompts don't work as a security boundary for coding agents
umangcirvix profile

Why approval prompts don't work as a security boundary for coding agents

#ai #security #mcp #devops
Comments
3 min read
5 permission rules every coding agent should ship with
umangcirvix profile

5 permission rules every coding agent should ship with

#ai #security #opensource #devops
Comments
4 min read
Stop the read-then-exfiltrate chain: session taint for AI agents
umangcirvix profile

Stop the read-then-exfiltrate chain: session taint for AI agents

#ai #security #opensource #mcp
1
Comments 1
4 min read
MCP Security: Where to Put the Authorization Boundary
umangcirvix profile

MCP Security: Where to Put the Authorization Boundary

#mcp #security #ai #agents
1
Comments 1
8 min read
AI Agent Authorization: Control What Agents Can Do
umangcirvix profile

AI Agent Authorization: Control What Agents Can Do

#ai #security #agents #llm
Comments
8 min read
Runtime Authorization for AI Agents: A Practical Guide
umangcirvix profile

Runtime Authorization for AI Agents: A Practical Guide

#ai #security #agents #opensource
Comments
9 min read
loading...