Writing about the identity layer
for autonomous AI.
Engineering deep-dives, security postmortems, and strategic takes from the AuthSec team. We publish what we wish we’d read when we started building this.
Trust Delegation: The Missing Piece in Multi-Team Identity
AuthSec lets you establish trust relationships between different administrative domains. Platform teams maintain the root of trust. Application teams receive delegated authority scoped to their namespaces, environments, or services—granular, time-bound, and cryptographically verifiable.
Secure Your AutoGen Agents With AuthSec Delegation Tokens
AutoGen's multi-agent conversations are powerful — but every registered function your agents call is a potential credential leak. authsec-autogen replaces static API keys with short-lived, RS256-signed delegation tokens so your AssistantAgent never touches a raw credential.
Secure Your CrewAI Agents With AuthSec Delegation Tokens
Static API keys in CrewAI tools give every agent a master key that never expires. AuthSec's authsec-crewai SDK replaces them with short-lived, RS256-signed delegation tokens — scoped per call, auditable, and dead on arrival if they leak.
MORE ESSAYS
From the archive.
Secure Your LlamaIndex RAG Pipeline With AuthSec Delegation Tokens
LlamaIndex ingestion pipelines reach into your most sensitive internal APIs. Static credentials make a leak catastrophic. AuthSec's authsec-llamaindex SDK replaces them with reader-native delegation tokens — short-lived, scoped, and stamped on every Document at ingestion time.
How We Secured Our AI Agent's API Calls Using AuthSec and LangChain
AI agents need to call real APIs on behalf of real users — but static API keys hand them a master key to everything. Here's how we replaced long-lived service tokens with AuthSec delegation tokens in LangChain tools, giving agents a bounded, traceable, short-lived identity for every downstream call.
Transfer Claude Code Sessions to OpenAI Codex and Gemini — One Command to Sync AI Agent Context
You hit Claude context limit mid-problem, switch to Codex, and spend 20 minutes re-explaining the same architecture. We built a small CLI that transfers your AI agent session — context, decisions, tradeoffs — between Claude Code, Codex, and Gemini in one command.
How to Install, Verify, and Use AuthSec Agent Shield
A complete step-by-step guide to getting AuthSec Agent Shield running on your machine — from building the binary to approving your first blocked command on your phone.
We Built a Firewall for AI Agents. Here's What Happened When Claude Code Tried to Delete Our Files.
You give Claude Code your terminal. You give Codex your codebase. You give OpenClaw your entire machine — and they can rm -rf your home directory or git push --force to production before you know it. AuthSec Agent Shield stops that at the OS level. Here's what happened when we put it head-to-head with Claude Code.
AuthSec for AI: Your Organization's Kill Switch for AI Agent Actions
AI agents can delete files, push code, query databases, scale infrastructure, and send customer emails with a single prompt. AuthSec adds a human approval layer before risky actions execute.
Building an Authenticated AI Gateway: How We Put OpenClaw Behind Enterprise SSO
AI assistants are powerful, but deploying one inside an organization without authentication is like leaving the front door wide open. Here's how we built authsec-openclaw — a Go reverse proxy that wraps OpenClaw with enterprise-grade SSO and gives every chat session a verified identity.
SCIM: The Enterprise Gateway Your AuthSec Deployment Needs
Enterprise customers won't buy your SaaS if they have to manually provision users. Every new hire means a ticket. Every termination means a security gap. AuthSec SCIM fixes this by automating user sync from Okta, Entra ID, and Google Workspace.
AuthSec: Open-Source Identity for AI Agents
Most identity products were designed around a very specific assumption—there's a human sitting in front of a browser. That model breaks down fast once you start shipping agents. AuthSec was built to fix that.
WRITTEN BY
The people building AuthSec.
Engineers, security folks, and a couple of product people who can’t stop writing about identity.
↳ NEWSLETTER
Get the next essay in your inbox.
One letter every other Tuesday. Production notes, no hype.