The Identity Navigator

Should you block AI agents? This episode answers it from the identity chair, and it starts with MCP authorization explained from zero. If you've heard "MCP authentication" and "OAuth" in the same breath all year and never quite connected them, this episode does it step by step.

We starts with OAuth 2.1 from scratch (what the roles are, what a token actually is, why it's not login), then MCP from scratch (what the Model Context Protocol is, local versus remote servers, why an AI agent is unlike any API client you'vesecured before), then walks the full MCP authorization handshake with the July 28, 2026 spec open: the 401, Protected Resource Metadata, Client ID Metadata Documents, PKCE, resource indicators, issuer validation, audience binding,step-up scopes, and the two rules that matter most. Then why this lands on the identity leader's desk, and the question the

CISO will actually ask: do we let agents in at all? We walk the block-or-admit debate, the August 2026 Ninth Circuit ruling in Amazon v. Perplexity, why the fraud bouncer can't tell a goodagent from an account-takeover bot, the screen-scraping-to-open-banking precedent, what Reg E and Reg Z do with "my agent did it," the managed-delegation posture (block anonymous agents, admit authenticated ones),and the two controls to turn up now, step-up authentication on anything that moves money and behavior baselining for the agent itself. Then gotchas, anti-patterns, and a Monday list. T

Why this episode now

OAuth has never been easier to learn, and MCP authorization guides exist, mostly written for developers building servers. This episode connects the two from the identity practitioner's chair: a six-minute OAuth refresher, MCP from zero, the handshake told as one scene (an intern walking up to room 412), what it changes about running an identity platform, the block-or-admit question for customer-facing agents, and why every change in the MCP spec this year was an OAuth lesson from years ago.

Thank you for listening

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

The Badge That Printed Itself: A SharePoint TokenForgery, Decoded Word by Word

One line from my morning threat intel digest: “CVE-2026-55040. Critical. CVSS 9.1. SharePoint on-prem authentication bypass, under active exploitation.” This episode stops and takes the whole sentence apart, from first principles, until it reads likeplain English. Then we walk the heist: how a stranger with no account and no password handed SharePoint a homemade admin token, and SharePoint said welcome back.

In this episode:

1. Decoding the line: CVE, CVSS, KEV, on-prem, auth bypass, PoC, and why an n-day means the attacker is racing your change board
2. What a token is, explained so the bug becomes obvious
3. The heist in four dominoes: alg “none”, the x5t pointer aimed at SharePoint’s own certificate, circular trust, and a signature that just says “AAAA”
4. Textbook Swiss cheese, and why four checks that share one assumption are really one check
5. Why forged tokens are quieter than stolen ones, and the three places you can still catch them
6. The out-of-the-loop identity platform: your MFA and conditional access were never beaten, they were never asked
7. Should you ever roll your own token validator? The steelman, the verdict, and the architecture I’d build instead

LinkedIn: ⁠https://www.linkedin.com/in/rohit-agnihotri⁠
Email: [email protected]

Three labs. Five breached companies. One testing vendor named Irregular.

This summer handed identity and security practitioners a masterclass in what happens when AI agents hit environments that weren't built for them. Not through sophisticated cyberattacks. Through configuration mistakes, weak passwords, and an optimizer that found a shortcut through Hugging Face's production infrastructure because the answer key wasn't in the sandbox.

In this episode, I walk through all three incidents - OpenAI's ExploitGym escape, Anthropic's 141,006-evaluation audit, and the invisible HTML comment that turns your developer's AI assistant into an insider threat.

Then I go somewhere the coverage hasn't: what the narrative around these incidents is doing to how practitioners make decisions, and why fear and FOMO are the same emotiondepending on which side of the paywall you're standing on.

Then we get practical. The identity air gap isn't about going back to 2003. It's about protecting the quality of your strongest signal, in a decision control plane that evaluates attributes like provenance, location, and device posture to produce a fidelity assessment. Agents introduced an attribute nobody was capturing. This episode is about capturing it.

Separate trust planes per environment. No credential valid in two worlds. Every crossing brokered, scoped, and provable end to end. Sever the path. Or make it provable.

Thank you for listening

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

Everyone is using AI. Almost nobody can prove what it’s doing. In this solo episode, we design a runtime AI governance platform from scratch on a whiteboard, and then meet PulseAI, the product IAM veteran Amit Masand built at IDMEXPRESS. With the numbers justifying the problem, an honest competitive map, who it’s not for, and a free AI-traffic risk assessment for listeners.

Links

• PulseAI by IDMEXPRESS: idmexpress.ai
• Company site: idmexpress.com)
• Amit Masand on LinkedIn: linkedin.com/in/amit-m-67b4692
• Contact: [email protected]
• Listener offer: mention this episode for a complimentary risk assessment of your AI traffic, no commitment.

Thank you for listening: Host's contacts

LinkedIn: ⁠https://www.linkedin.com/in/rohit-agnihotri⁠
Email: [email protected]

Previously on The Identity Navigator (recommendedlistening before this episode): Episode 40: SCIM - The protocol that finally made provisioning grow up

Every app you onboard forces a quiet architecture decision most teams never make explicitly: does provisioning belong to your IGA platform (OIM, SailPoint IIQ, Saviynt) or your IdP (Entra, Okta)? In this episode, I walk through the five questions that settle it - before the app lands in your ecosystem, not after.

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

0-40% of SaaS accounts in the mid-2000s belonged to people who'd already left the company. Ghost accounts, quietly sitting there — and that sadly hasn't changed much today.

This episode: where SCIM came from, what it actually solves, and what it still doesn't.

• Core RFCs:
  RFC 7642 — SCIM: Definitions, Overview, Concepts, and Requirements
  RFC 7643 — SCIM: Core Schema
  RFC 7644 — SCIM: Protocol

• Key Terms Covered:
  SCIM 2.0 — System for Cross-domain Identity Management
  JML — Joiner-Mover-Leaver lifecycle
  IGA — Identity Governance and Administration
  IDP — Identity Provider (Okta, Entra ID, Ping, JumpCloud, OneLogin)
  JIT — Just-In-Time provisioning
  SPML — Service Provisioning Markup Language (the predecessor that didn't make it)
  PatchOp — SCIM's structured partial update format
  ServiceProviderConfig — SCIM's capability discovery endpoint

• Tools Mentioned:
  Identity Providers: Okta, Microsoft Entra ID, Ping Identity, JumpCloud, OneLogin, CyberArk Identity
  IGA Platforms: SailPoint, Saviynt, IBM Security Verify Governance, Omada Identity
  Key SCIM-enabled Apps: Slack, Salesforce, GitHub, GitLab, Workday, ServiceNow, Snowflake, Zoom, Box, Atlassian
  
• SCIM Validator:
  scimvalidator.net — for testing SCIM endpoint compliance

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

The Complete Token Story

Let me make you a promise: by the end of this episode, no one will be able to casually brush off the word “tokens” in front of you ever again.

OAuth, Tokens, JWT, IAM, Identity, API Security, Zero Trust, Security Architecture, CISO, Engineering Leadership

Thank you for listening

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

In this episode of The Identity Navigator, Rohit explores workload identity from first principles: what it is, why it matters, and how it is reshaping modern identity security for microservices, Kubernetes, CI/CD pipelines, cloud workloads, and AI agents.

The discussion compares workload identity with non-human identity management, API keys, Vault, and cloud-native identity services, then goes deeper into SPIFFE, SPIRE, federation, and the future of identity in agentic systems. If you want a clear, practical explanation of workload identity and why it is becoming essential in zero trust architectures, this episode is for you.

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

In this episode we explore together the “uncanny valley” of identity categorization, where taxonomy looks mature on paper but doesn’t meaningfully change controls in practice.

We talk about Taxonomic Debt and control plane behind them. From service accounts and workload identities to bots, technical users, and AI agents, we argues that the real unit ofgovernance is not the label, it’s the behavior: interactive ornon-interactive, ephemeral or persistent, privileged or not.

If your IAM program has ever felt like governance theater, this episode will help you see why.

You’ll walk away with a clearer way to categorize identities by how they behave, how they’re secured, and how the system should react when they drift from expectations.

Thank you for listening

LinkedIn: https://www.linkedin.com/in/rohit-agnihotri

Email: [email protected]

In late November 2022, a little-known app called Lensa went from obscurity to everywhere. Celebrities, influencers, and millions of ordinary people uploaded their selfies and got back stunning AI-generated portraits, fantasy warriors, Renaissance paintings, cyberpunk heroes, and more. The app’s viral “MagicAvatars” feature launched in late November and propelled Lensa to the No. 1 spot on the iOS App Store’s Photo & Video charts.

At the center of thefrenzy was a simple pitch: pay a few dollars, upload a handful of selfies, and watch AI turn you into art. But the backlash arrived fast. Critics flagged hypersexualized outputs for women, artist concerns over training data and styleappropriation, and privacy questions about what users were actually agreeing to when they uploaded their faces. Reporting at the time noted that Prisma Labs’ terms allowed the app to use user content to operate or improve the service, and that the company updated its privacy policy in December 2022 amid thecontroversy. 

This episode is a story about virality, timing, and the dark incentives hiding inside consumer AI. It’s about how a polished interface, an irresistible social loop, and a moment of cultural hype can turn into an extraordinary revenue machine.

TechCrunch reported that Lensa generated more than $70 million from the app in November 2022 alone, with Sensor Tower data showing the app’s downloads jumping to 1.6 million in November, up 631% from October.

But the bigger question is not whether Lensa was clever. It’s what its success reveals about the AI era: speed can outrun ethics, product can outrun governance, and ordinary users often surrender far more than they realize in exchange for convenience and novelty.

“The following represents my analysis and commentary based on publicly available information and reporting.”