# Decoding Mastro: AI Threat Modeling

URL: https://securitytable.ai/decoding-mastro-ai-threat-modeling/
Published: 2025-09-10
Duration: 2946 seconds
Season: 3
Episode: 13
Hosts: Chris Romeo, Izar Tarandach, Matt Coles

## Show notes and chapters

We’re discussing the article, “Agentic AI Threat Modeling Framework: Maestro published back in February of this year on the Cloud Security Alliance blog. We discuss the various layers, patterns, and threats outlined in the framework, comparing it to existing methodologies like STRIDE and PASTA, and evaluate Maestro's structure, its potential complexity for developers, and its overall practicality and usefulness in the threat modeling arena. Listen along as we unravel the intricacies of the framework and share our candid thoughts on its strengths and weaknesses.

[Agentic AI Threat Modeling Framework Maestro](https://cloudsecurityalliance.org/blog/2025/02/06/agentic-ai-threat-modeling-framework-maestro)

  
  

Follow AI Security Table:

➜ Home: [https://securitytable.ai/](https://securitytable.ai/)

➜ X: [https://x.com/AISecurityTable](https://x.com/AISecurityTable)

➜ LinkedIn: [https://www.linkedin.com/company/ai-security-table/](https://www.linkedin.com/company/ai-security-table/)

➜ YouTube: [https://www.youtube.com/@AISecurityTable](https://www.youtube.com/@AISecurityTable)

## Transcript

**Chris Romeo (0:08)**: Hey folks, welcome to another episode of the Security Table. This is Chris Romeo joined by Izar Tarendash and Matt Coles, the keepers of the Security Table. Not sure what that really entails, but sounds mythical and like it was derived from a fantasy. Novel somewhere.

**Izar Tarandach (0:26)**: It, it, it's in the back room right here and we just roll it out once a week.

**Chris Romeo (0:31)**: That's how we keep it.

**Matt Coles (0:32)**: Where, where, where's our sword? Where's our swords and shields?

**Chris Romeo (0:36)**: Uh, I thought we were, yeah, I, it's in my contract that I will specifically be issued a sword that is sharp.

**Izar Tarandach (0:42)**: We Knights of Security Table, we dance whenever they boo with the routines.

**Chris Romeo (0:47)**: We have a theme song.

**Izar Tarandach (0:49)**: All right.

**Chris Romeo (0:50)**: We're just gonna need, can you record that?

**Izar Tarandach (0:52)**: Actually, we stole that one, but yeah.

**Chris Romeo (0:54)**: Yeah. Could you just record a version of that? Or I can just, I can just generate one with my AI voice.

**Izar Tarandach (1:00)**: Did you guys see the, the, the OWASP songs in the Leaders channel generated by that, uh, AI engine?

**Matt Coles (1:06)**: I don't wanna know.

**Izar Tarandach (1:07)**: There is some pretty good stuff there. Really? Really. Did they let it analyze the Leaders list so that it could find the equivalent of, uh, of, uh, listening to Eye of the Tiger on the way to the gym? Like you really wanna get that thing secure. No, really, some good stuff there.

**Matt Coles (1:26)**: By reflection, there's also some really bad stuff in there.

**Izar Tarandach (1:32)**: I'm not going to say it's a zero-sum game, but it's—

**Chris Romeo (1:37)**: Are we talking about the OWASP News channel or the songs?

**Matt Coles (1:41)**: I was talking about the songs.

**Izar Tarandach (1:44)**: Yeah, the songs. Talk to me about the songs. Totally.

**Chris Romeo (1:46)**: Totally. Yeah.

**Izar Tarandach (1:47)**: Yeah.

**Chris Romeo (1:48)**: Yeah.

**Izar Tarandach (1:48)**: Not a word. Hey, which reminds me, it's, uh, OWASP, uh, board of directors election time again.

**Chris Romeo (1:55)**: It is.

**Matt Coles (1:56)**: Yeah.

**Izar Tarandach (1:57)**: Matt, did you?

**Matt Coles (1:59)**: No. Matt, did you?

**Izar Tarandach (2:02)**: Well, now you can't anymore because it's over.

**Matt Coles (2:05)**: Oh, it closed?

**Izar Tarandach (2:06)**: It's locked in.

**Chris Romeo (2:07)**: No.

**Izar Tarandach (2:08)**: But Matt, did you?

**Matt Coles (2:09)**: No.

**Izar Tarandach (2:10)**: Matt, did you?

**Chris Romeo (2:12)**: This is getting weird.

**Matt Coles (2:15)**: And if, if somebody wants, if somebody wants to, uh, put my name in with, I'm looking for my mic. I can't, I can't, I, I do, I like you too much for that, but I'm sure it will be a great slate of candidates this year.

**Chris Romeo (2:33)**: Yeah, me too.

**Matt Coles (2:36)**: We sound so committed.

**Chris Romeo (2:37)**: Well, it's early in the morning here. The eastern coast of the United States.

**Izar Tarandach (2:43)**: Okay.

**Chris Romeo (2:44)**: We get, we get, we should dive into this because this is probably gonna take a little while.

**Matt Coles (2:49)**: Oh yeah.

**Chris Romeo (2:50)**: The topic of the day, and we'll put this into the show notes, is Agentic AI Threat Modeling Framework, Maestro. And this was published on February 6th, 2025. I cannot, see this name without thinking about the character from Seinfeld, the Maestro.

**Matt Coles (3:12)**: Yep. Or Sorcerer's Apprentice.

**Chris Romeo (3:16)**: That's even better. That's, that's even more unique of a pull there, right? Okay, so Maestro, Maestro. It's, it's a, uh, AI. Yeah, AI. Well, I mean, every time we have another framework, and I'm gonna save my, I did a little bit of digging into this, and I have a conclusion that I think I can draw, but I'm gonna kind of let it be more of a crescendo, trying to work in Maestro-related terms. I don't have a lot of them to work with, but crescendo is one of them. Okay, so starts out with a comparison of existing threat modeling frameworks. And every time I see this list, everybody uses it. I'm like, do you know anyone that uses Octave, Trike, or Vast? Anyone?

**Matt Coles (4:08)**: I know people who use Vast.

**Izar Tarandach (4:10)**: There's a vast community that got on their trike and just Octaveled out of here.

**Chris Romeo (4:16)**: Okay. So, all right. Octave and Trike. Do you know of, I don't, I do not know of anyone who's using these.

**Matt Coles (4:22)**: Octave is an old ancient system that is probably used in some regulated environments and and, uh, particular industries, probably just because they haven't changed in X number of years.

**Izar Tarandach (4:34)**: That, that sounds like a promo for Stargate. It's an ancient system that's used only on Kulos.

**Matt Coles (4:41)**: Well, remember, OCTAVE is a giant acronym, right? But it's not a threat modeling acronym.

**Chris Romeo (4:47)**: Yeah. It says organizational risk.

**Matt Coles (4:50)**: Yep.

**Chris Romeo (4:50)**: That seems like something that's very painful.

**Izar Tarandach (4:53)**: But, but, but let's be positive here. It does not say dread and it does not say attack trees.

**Matt Coles (5:01)**: Or CVSS.

**Izar Tarandach (5:02)**: Or CVSS. So let's be positive here.

**Chris Romeo (5:05)**: True. CVSS, why does that always show up as a threat modeling framework?

**Matt Coles (5:10)**: Just for completeness, Octave, there's operationally critical threat asset and vulnerability evaluation.

**Izar Tarandach (5:17)**: Yeah, sure, why not?

**Matt Coles (5:19)**: Which is interesting because that sounds a lot like The acronym for pasta.

**Chris Romeo (5:24)**: Well, so Stride, Pasta, Linden are the first 3 on the list, which I was just going to let those be, um, as you know, it's, it's worth when you're building something to compare it to what exists in the, in the, uh, known universe. Um, what you would— I think, I think what you would expect for strengths and weaknesses for Stride, Pasta, and Linden against what has been created here is Maestro, you know, basically saying Stride's great, but it's too simple. I'll skip over the PaSTA one. And Linden is about privacy, which privacy has connective, connective, uh, pieces into agentic AI, but Linden wasn't created for AI. It was created for a kind of base layer of, of privacy, right?

**Matt Coles (6:10)**: Yeah, but more importantly, Linden and, and Stride especially, and to some extent PaSTA as well. Are general purpose threat modeling approaches, right? They can be applied, they can be applied to any type of threat modeling, whether you're doing, whether you're doing process, you're doing systems, you're doing privacy, you're doing security, you want to do both at the same time, right? You can apply these in a general purpose way. Uh, if you're doing space threat modeling, you have space systems, you're doing aeronautics, you're doing trains, ports, whatever, you can apply these approaches. They work better for some things than others, obviously, because of the awful literature that's available, but the, the approach of threat modeling that they follow, and really STRIDE and Lindem are not really methodologies, more so they're threat categories, right? PASTA is actually a methodology and not a threat category, set of threat categories, right? So, but these are general purpose systems that are methodologies that can be used. You know, pair them with, say, STRIDE for element or STRIDE for interaction or the 4-question framework as a methodology for applying the threats, they're general purpose. I think what CSA, well, what CSA did, not what I think, but what they did was they created a domain-specific threat modeling approach, kind of like a domain-specific language.

**Izar Tarandach (7:31)**: So here's the thing. We, we, we, we keep moving between those terms, and, and I think that you addressed it perfectly. If you look at Stride, Stride is a categorization of threads. If you look at Stride per LM, sorry, Stride per element.

**Matt Coles (7:47)**: Not per LLM.

**Izar Tarandach (7:49)**: Let's not go there. No, actually, let's in a bit. But Stride per element, then it's the application of Stride in a certain way over elements of a system. So one is talking about a categorization. The other is talking about how to apply that categorization. So we are jumping from one thing to another thing. You have a catalog of things and you have how to apply the catalog of things. And somehow we ended up calling both of those things methodologies, which I believe is wrong.

**Matt Coles (8:27)**: Yeah.

**Izar Tarandach (8:28)**: Now we come to Maestro. And I look at my astro and I'm asking myself, what is it? Because it confuses me.

**Matt Coles (8:40)**: Well, so—

**Izar Tarandach (8:41)**: Is it a catalog? Is it a methodology? Is it both? Is it none?

**Matt Coles (8:45)**: It's kind of both.

**Izar Tarandach (8:45)**: Does it come in green?

**Matt Coles (8:47)**: It's kind of both. And actually, so then they, if you read the description, I guess we could put the link in the, into the show notes. They describe a set of threats and a methodology for applying those threats or identifying those threats across an agentic system. I do want to call out something of interest. I don't know if anyone's noticed this, but Maestro is an agentic AI threat modeling methodology and, and set of threats. There's no general GenAI or LLM threat modeling specific methodology.

**Izar Tarandach (9:23)**: So there's the thing. Maestro comes as a layer reference, right?

**Matt Coles (9:30)**: Yep.

**Izar Tarandach (9:31)**: And one of those layers is the foundation models.

**Matt Coles (9:36)**: Yes. Layer 1, in fact.

**Izar Tarandach (9:38)**: Yeah. Now, let me step that back for a second. We grew up in a world where you said 7 layers, you immediately thought about one thing. And I'm not thinking—

**Chris Romeo (9:52)**: It was dip. It was like a dip, 7-layer dip. Yeah, yeah. You put your chips in, boom.

**Izar Tarandach (9:58)**: And I'm not thinking ground meat, marshmallow, and Monica. What I'm thinking is we will think about the TCP/IP layer thing, right? So you have the frame, and inside the frame you have the TCP/IP, and then you have the physical layer, right? With the physical layer, yeah. Right? So things got encapsulated and you couldn't close your eyes and see that layer thing actually developing and understand from that point of view. I don't get the layers of mastery.

**Matt Coles (10:38)**: Well, so let's go through them, right? So I'm looking at the chart here. Layer 1 is foundation models. Right? We know what those are, right? We think we know what those are. Layer 2 is data operations. So, right off the bat, I think what you're highlighting is OSI was a, a layer, it was a layered approach because it was a structure and a set of encapsulations, as you mentioned, right? Mm-hmm. The physical layer was a communication layer. Now on top of that, there was a, there were frames and there were packets and there was a, this sort of grouping of things that that ended up with the application at the top having its own understanding of the bits. Whereas it seems that Maestro is building sort of a collection of services and how they relate to each other.

**Izar Tarandach (11:23)**: Or structures, even structures.

**Matt Coles (11:25)**: But I guess if you look at it as like a, as a high-level diagram or, or as a, a functional model, right? You have, you have a foundation model as a, grouping of, of things, and then you have data operations, which is a function of those models. And then there's agent frameworks, which are functions of data operations and models, right?

**Chris Romeo (11:44)**: I don't think that was their— I don't think that was the intent here. No, it's, it's not an OSI encapsulation thing where they all stack and they're, they're fit inside of the box of the previous one in more detail, right?

**Izar Tarandach (11:55)**: And, and these are not layers because they, they don't base themselves on the previous one. Okay, you can have an agent framework without any concern whatsoever for foundational models. Most agent frameworks, by the way, you just say which model we want to use. You don't have to go thinking about all the things that could go wrong with a foundational model.

**Chris Romeo (12:16)**: But I mean, our solid architecture, solid architecture is a layered approach, right? From top to bottom where we do understand, or at least have the ability to understand you know, if, if the three of us decided we were going to build an agentic AI solution, we would 100% look at the foundational models and, and consider what are the potential threats that exist in us building something on top of that. So my one thought on this is I, I kind of, I had an initial reaction like you did, Isar, and then I kind of stared at it for a while and I read some of the descriptions and I realized it does kind of build on top of the previous layers towards that agent being available, basically, in the ecosystem. The only one that I don't like is layer 6 security and compliance. I was just looking at that.

**Matt Coles (13:08)**: Like, why is that so far up?

**Chris Romeo (13:10)**: Yeah, why do I— I don't care about security till I get to layer 6. Like, you got to pull security and compliance off, turn it sideways, and let it ride across all of the layers.

**Izar Tarandach (13:18)**: Exactly, because if you're waiting until layer 6 to do security and compliance, What's happening on 1 to 5?

**Chris Romeo (13:24)**: Yeah.

**Matt Coles (13:25)**: Well, by that same measure though, evaluation and observability, and even to some extent deployment and infrastructure is sort of this cross-cutting thing as well, right? So really your foundation models are things, your operations are things, your frameworks are things, your agent ecosystem are things. And then those 4, 5, and 6 are really cross-cutting across all of those. You deploy foundation models, you evaluate data operations, and you need security compliance for all of those. Mm-hmm. Right? So it's really a box.

**Izar Tarandach (13:59)**: Yeah, but, but I don't, I don't, I'm sorry. I, I keep looking at the layers as if it's just tasty lasagna. What actually, what you get is an amazing spaghetti.

**Matt Coles (14:10)**: Oh, remember Spaghetti Code? Remember when we loved Spaghetti Code?

**Chris Romeo (14:14)**: Oh, but, but don't write code like that.

**Izar Tarandach (14:16)**: What are you talking about? No, no, no, no, no, no. We vibe through that. So don't worry. The agents are going to fix our spaghetti code.

**Chris Romeo (14:21)**: The agent creates the—

**Matt Coles (14:22)**: I was gonna say, do agents create spaghetti code? Can you ask them to?

**Chris Romeo (14:27)**: I bet you that's what they do by default.

**Izar Tarandach (14:28)**: It depends on the foundational model.

**Chris Romeo (14:32)**: Nice fake Italian accent there. Okay.

**Matt Coles (14:35)**: So for hate mail, send it to—

**Chris Romeo (14:37)**: send it to Isar.

**Izar Tarandach (14:39)**: Hey, hey, hey, hey, hey, hey. I hate, I hate when people say that, that making an accent is offensive because it's not, it's just showing that you love that. And I love that. So there's a lot of love in Italian food. I do. I do. Okay.

**Chris Romeo (14:57)**: So let me bounce a big picture idea off you that hit me every time I look at a new framework like this, my immediate reaction is. Everything is STRIDE. No.

**Izar Tarandach (15:12)**: Yes.

**Chris Romeo (15:13)**: But I had, I had an, I had a, I had ChatGPT help me to map which STRIDE categories apply to each of the threats in Maestro. And so one of the, so one of the challenges in the way he stacked these things up is the threats duplicate across the layers.

**Matt Coles (15:29)**: Yeah, they do.

**Chris Romeo (15:30)**: They step. All of that to say, like, what I love about STRIDE is it's simple. I can explain it to somebody in 5 minutes. I couldn't explain this to somebody in 5 minutes.

**Izar Tarandach (15:41)**: I couldn't even really explain it. Well, if they understand it in 5 minutes, it's a completely different discussion. But yeah, you can explain to somebody.

**Matt Coles (15:46)**: Sure.

**Chris Romeo (15:46)**: But you can explain it in 5 minutes. And so when I've start— when I started going through this, I'm like, it's just compli— it's, it's, it's not simple. Like simple is what makes things possible for people to use.

**Izar Tarandach (16:00)**: It's too complex. It's not that it's not simple. I feel like it's shoehorned. Like, we have this imperfect solution. Uh-oh. Not solution. We have this imperfect thing that we can use, that we already know that has been tested over the years, that has a clear approach. To look at those things that we have been saying for a long time, these are not new things.

**Chris Romeo (16:27)**: These are just— So, I can—

**Izar Tarandach (16:28)**: new flavors of old things.

**Matt Coles (16:30)**: So let me tell you, this is, this, so I'm, I'm, I'm totally with both of you. And Stride is so easy to use because it's a general purpose thing again. You know what, you know what Maestro really is? Maestro is really a threat library.

**Izar Tarandach (16:46)**: Yeah, exactly. It's another catalog.

**Matt Coles (16:49)**: It's a catalog of specific threats per layer, right? So specific threats per domain, per thing, right? And, and that will, like, right, those things are encoded in layers. And, and so you're going to look for this set of threats when you do foundational models and only this set of threats, right? Whereas Stride is, you're going to look across your, your acronym, right, of what can go wrong in that layer. So why couldn't you have applied Stride to the, like, take the late, take the layers out. Take, pull the layers out and make them its own thing and use that as a methodology for applying Stride or Linten directly.

**Chris Romeo (17:31)**: Yep. Or both.

**Matt Coles (17:32)**: It would have worked.

**Chris Romeo (17:35)**: Yeah. And that's what my ChatGPT analysis determined.

**Matt Coles (17:40)**: I'm better than ChatGPT. I mean, sorry. My thing is that equivalent to ChatGPT.

**Izar Tarandach (17:44)**: Then at some point Maestro jumps into cross-layer threats, right? And what's cross-layer threats? It's supply chain attacks. It's lateral movement. It's privilege escalation, data leakage, and goal misalignment cascades, which it's a, it's the one that I look at and say, you know what, cool. That's, that's something that might be new, but the other 4, I'm sorry. Why? Why, why, why complicate things that can at this moment in time be made simpler? Right?

**Matt Coles (18:16)**: I, I have a feeling that's because of interactions. So one of the things about the layers is the layers are, the layers are described as these self-contained things. And then the cross, which is, which is interesting in itself. 'Cause you look at the foundation models, for instance, it's, if you look at the foundation model as a, as a layer that's self-contained, like Flatworld, you can't go above and below it. There's stuff that happens at that layer in that 2D layer, ignoring the 3D part. The cross-cutting is basically recognizing that there is a third dimension to the, to this, to this set of things. And there's a bleed of, a bleed across these layers.

**Izar Tarandach (18:58)**: If you need to create a separate class of things that are cross-layer to show that there is a justification to the layers, then I would offer that the problem is not being explained right.

**Chris Romeo (19:12)**: Okay.

**Izar Tarandach (19:12)**: Because take prompt injection, for example, that we love to bang on. Foundational models, they are what they are, right? They have their guardrails. They come with their stuff. Not everybody who's going to write an agent right now, for whatever that means, is going to look at a foundational model and say, you have to retrain this thing to be much more robust against prompt injection. No. They're going to take the, the, the, the Kellogg's box off the shelf, use it as it is, and slap some stuff like Llama Firewall on top of it and say, okay, I took care of prompt injection. Or I said, ignore anything that looks like ignore your previous instructions. It's not like they're going to go down into the layer of the foundational model and say, what are the things that I have to worry about here? And what are my mitigations? Because they're not going to affect those mitigations. So why focus there?

**Matt Coles (20:09)**: So, so I've got something else to throw at me for you to, to, to noodle over and, and, and bring that fire, Isaur. I, I think I can, I'm, I'm gonna inflame the, the rage that you have here.

**Izar Tarandach (20:20)**: I'm not raging. Yeah.

**Matt Coles (20:21)**: 3, 3 things I, 3 things I just noticed as I reread this. First, to Chris's point, look at layer 6 in the full description if you're reading the article.

**Izar Tarandach (20:34)**: I am looking at it.

**Matt Coles (20:35)**: Layer 6, security and complaints, parentheses, something.

**Chris Romeo (20:40)**: Vertical layer.

**Matt Coles (20:41)**: Vertical layer. So—

**Izar Tarandach (20:43)**: Cuts across all of the layers. So this one, this one is the drunken layer. Everybody's like, no, this one goes 90 degrees.

**Matt Coles (20:53)**: And then in Layer 3, agent frameworks, take a look at bullet, sub-bullet number 4.

**Izar Tarandach (21:00)**: You're making me count early in the morning.

**Matt Coles (21:02)**: I am, yes.

**Izar Tarandach (21:03)**: Supply chain attacks targeting the AI framework's dependencies, compromising software before delivery and distribution, resulting in compromised AI.

**Matt Coles (21:11)**: So the layers— So the layers already have the cross-cutting attacks built into them. So why have a separate category of cross-cutting attacks? All right. Last, last point. I dare you find prompt injection in here?

**Izar Tarandach (21:30)**: I'm sure that it's there. I'm sure that it's there but with some fancy name unconstrained conversational autonomy okay a conversational AI agent that can process and respond to a wide range of inputs without tight constraints so what is this wait—

**Chris Romeo (21:50)**: Wait!

**Matt Coles (21:50)**: What now?!

**Izar Tarandach (21:50)**: So let me ask: "Wait!" Okay...

**Chris Romeo (21:52)**: right.

**Izar Tarandach (21:52)**: Right.

**Chris Romeo (21:52)**: Right. Right.

**Izar Tarandach (21:53)**: Right. Right. Oh no, no, actually, go.

**Matt Coles (21:55)**: What is this? What is unconstrained conservational autonomy? Where did this come from? Uh, this is a pattern. So this maestro, this maestro also includes patterns in addition to the layer, in addition to the layers. So now it's even more confusing.

**Izar Tarandach (22:15)**: Yeah.

**Matt Coles (22:15)**: Where do the patterns fit into the layers?

**Izar Tarandach (22:17)**: And now that we are looking at patterns, I'm, I'm going to go with something That really, to me, was very vexing. When we get to the mitigations, okay, so the mitigation for unconstrained conversational autonomy is robust input validation and safety filters designed specifically for the conversational AI use case. I'm a developer. I just read that sentence. Either I am going to get up and go breathe some fresh air, or I'm going to put this into ChatGPT and say, how do I do this? Because it's not practical.

**Matt Coles (23:04)**: It's jumbled is what it is. It's not practical because you, so how would you apply it? I guess that maybe that's the, that's the question we need to ask is how do you apply Maestro? So they have the layers, and then they have the patterns, and then they have other things. And some of those layers describe threats that are crosscutting, and some, and there's a separate category across those layers of crosscutting things. So how do you apply this?

**Izar Tarandach (23:35)**: I, I, I think that at this level, at this time in, in, in the life of the, the whole thing, and by whole thing, I say threat modeling AI. Not Maestro specifically. It might be argued that there is value by itself in pointing at the problem simply because we don't have solution answers for every single problem, right? I'm not aware of something that works 100% for prompt injection. And I have said 1,000 times, if I have said once, I don't think that prompt injection is a security problem. I think that The things that prompt injection enable are a security problem. But prompt injection itself, I think that is just inherent to the fact that the way that these things work, right? But I, as a threat modeler, as someone who tried to develop a methodology, as someone who tries to be as much as I can educational to developers, to just throw a sentence, a generic sentence. It feels like somebody coming in and saying, there is something broken in the car. And I go back to them and say, well, I fixed something in the car. You're not giving me a defined problem and I'm not giving you a defined solution, but both of us are happy because you told me that there's a problem and I told you that I solved something. So it, it, it's not the actionable part is flimsy.

**Matt Coles (25:13)**: I, I want to object to that a little bit, not because I don't think it's flimsy, but not for that reason. At least my show sort of brings together multiple things and really doesn't do a great, well, probably doesn't do a great job about doing all of those things well, if they had separated threats to agents and the mitigations for those threats as one group and how to look at agents and their frameworks and models to perform threat modeling and done that very well, this might have been a different conversation. But I think what you're highlighting is that they jumbled a lot of things together and, and didn't do all of it very well. Right? There's, there's a lot of information here. I mean, the, the threats per layer, those on our own are probably valuable.

**Izar Tarandach (26:13)**: Yeah.

**Matt Coles (26:13)**: The mitigations to those threats maybe needs work.

**Izar Tarandach (26:18)**: Indeed.

**Chris Romeo (26:20)**: Right.

**Matt Coles (26:20)**: Now, I guess one One thing that's probably important that hasn't, isn't described here, isn't discussed here, at least as far as I can see, uh, is the nature of agents as human equivalent processes.

**Chris Romeo (26:39)**: Isn't there a human in the loop mentioned here in the patterns?

**Matt Coles (26:42)**: There, there might be, but I'm not talking about human in the loop. I'm talking about agents are human.

**Izar Tarandach (26:47)**: Yes.

**Matt Coles (26:48)**: Effectively, from a purpose of a system, an agent is effectively like a human. When a human calls a function, when a human goes to a website and makes a function call and it, you know, through an interface, an agent can do the exact same thing. And it does it.

**Izar Tarandach (27:04)**: Identity with properties and— Right.

**Matt Coles (27:08)**: So anything that applies to humans applies to agents in terms of their function and their operation. How they're, how they're built, how they're deployed, how they're, how they're called. Maybe you don't call a human the same way that you call an agent, but so there's some very specific things around agents that are not human equivalents, but everything that's human equivalent in an agent and its operation or function should be able to use any of the human equivalent models that we already have.

**Izar Tarandach (27:43)**: I may be oversimplifying things, but for me, it's more like you go into a restaurant, you have a menu, and you have a chef, right? The chef is capable of doing a lot more. And the menu is just saying, well, in this restaurant, we are limiting the chef to doing these specific things.

**Matt Coles (28:01)**: That's what Short Library does, right? That's it on artificial limits.

**Izar Tarandach (28:06)**: I mean, for me, an agent would be the application of the menu on top of the chef.

**Chris Romeo (28:10)**: Oh, sure.

**Izar Tarandach (28:10)**: So an agent would be limiting and channeling an LLM to do these specific things, or, or, sorry, limiting and channeling the, the ability of an LLM to interpret things and create some seemingly appropriate output based on that input to these specific things, right? So one of the things that really calls out to me here on the, on Maestro and the layers and the mitigations and all that good stuff. Is that, and this may be a failure of mine, not of Maestro, but it's how I read the thing and the red light that comes on for me, is that in the middle of all the mitigations, in the middle of the generic approach to the mitigations and all that, that there's a jumble of personas in here that to me is very damaging to the whole effort. Not everybody who's going to apply this to an adjacent system is going to be able to, A, understand, B, evaluate, and C, perform the mitigations that are suggested at all levels. So this is going to generate a list of findings that will lead people to say, this is not relevant to me, I have no idea what this is, and I am buying this canned, so somebody else's responsibility here. And I'm not sure how, in its present form, without calling out that specific fact that not everybody is able to do the same thing at all the layers. I don't know if it's, if that's good, if that's bad, if it's something that should be better addressed, if it's Do we need Maestro for people who create foundational models? Do we need Maestro for people who create agents? Do we need to channel these things, separate these things? You understand my discomfort here?

**Chris Romeo (30:17)**: When you said personas, it's missing personas. Did you mean like personas of who people are going to use it or personas that are part of the threat library?

**Izar Tarandach (30:30)**: Personas of who's going to use it, because my understanding is that most of the time, the person who's just using framework like Lang-whatever, LangChain, Lang, the many Langs out there to build their client, right?

**Chris Romeo (30:49)**: Yeah.

**Izar Tarandach (30:49)**: It's not the same person who's training the foundational model. You have a data scientist doing the model and you have a developer doing the agent. And now you get this list of findings and this cross stuff and all that good stuff. And either each one of them is going to look at things in their appropriate way, or each one of them is going to say, this 90% of this thing doesn't touch me. So what's the, what's the value of this?

**Matt Coles (31:22)**: Let's go back to the Threat Modeling Manifesto. Do you do this by yourself or do you do this with a group?

**Chris Romeo (31:28)**: Yes, but. No, no.

**Izar Tarandach (31:29)**: Yes, but.

**Matt Coles (31:31)**: Yes, but.

**Izar Tarandach (31:32)**: Yes, but. My point is that most of the times, the people who can look at all these things in one organization are just not there.

**Matt Coles (31:42)**: Well, that's true. You're not buying the foundational model from yourself, right? Exactly. You get it from somewhere else.

**Chris Romeo (31:48)**: I'm laughing at what Izar just said. The people who could evaluate all these things are not there.

**Izar Tarandach (31:53)**: Yeah.

**Chris Romeo (31:53)**: They're not there. How are they the people that could evaluate all the things?

**Matt Coles (31:56)**: There are people that exist.

**Chris Romeo (31:57)**: It's a killer logic.

**Izar Tarandach (31:59)**: There are no, there are people who could be there, but they are not because, because there are other places.

**Matt Coles (32:04)**: There are other places. There are other places doing other things.

**Izar Tarandach (32:07)**: They're elsewhere. They exist, but not here. They are in a different plane of existence.

**Matt Coles (32:13)**: We think that they exist or they're all in his head.

**Chris Romeo (32:16)**: Schrödinger's people who could be doing AI agentic threat modeling.

**Izar Tarandach (32:21)**: No, it's, it's, it's, it's chess, chess higher. You, you only see their smiles.

**Matt Coles (32:26)**: Or, or it's, or it's, or it's 4D, it's 4D chess.

**Izar Tarandach (32:31)**: Yeah.

**Matt Coles (32:31)**: Uh, 4D laser chess. Uh, so I have a, I have a problem. I, I, I, I think I'm, I think I'm starting to understand the, uh, the mess, the spaghetti that is in his accent from earlier.

**Chris Romeo (32:44)**: That's very dangerous.

**Izar Tarandach (32:44)**: Explain to me then, please.

**Matt Coles (32:46)**: Please tell me why there's patterns with threats and there are layers with threats, but they're not the same threats. There's no connection between the threats on the patterns and the threats in the layers.

**Izar Tarandach (33:04)**: Wait, that came from my hand?

**Matt Coles (33:11)**: Uh, if you look at the threats that are in the patterns and you look for those threats in the layers, they don't match in many cases. In most cases, they don't match.

**Chris Romeo (33:21)**: I see what you're saying.

**Izar Tarandach (33:22)**: I heard you like threats. I heard you like threats, so I put threats into your patterns.

**Matt Coles (33:26)**: Well, so, as an example, prompt injection is in unconstrained conversational autonomy. It's not in any of the layers. Jailbreaking is also there. It's not in any of the layers. Unless it's redescribed as something, unrecognizable. It's not there. If you look at communication channel attacks on multi-pattern agents, or you look at compromise of a higher-level agent to control subordinates for the hierarchical agent pattern, where is that threat in the layer?

**Izar Tarandach (33:57)**: Well, to, to, to be, to be fair, in the page that we are looking at, it says that this section provides some examples of agentic architecture patterns and the risks associated with them. So, It's not exhaustive.

**Matt Coles (34:08)**: Sure. But why are they not in the layers? But why are they not in the layers? Because the layers are exhaustive.

**Izar Tarandach (34:15)**: Yes.

**Chris Romeo (34:16)**: So this is just an organizational problem that this document could have been organized to make it more simple to understand, I think is what you're ultimately trying to say.

**Matt Coles (34:25)**: Yeah. Either they choose the layers or they choose the patterns because they're building a threat library there, but they're building two distinct threat libraries.

**Chris Romeo (34:35)**: Right now. I see. Yeah.

**Izar Tarandach (34:36)**: So if we look at this thing—

**Chris Romeo (34:37)**: And there's duplication. And amongst the layers, there's duplication. So that, that makes the, it makes it even more challenging because the same threat appears in multiple spots.

**Matt Coles (34:46)**: Well, and then—

**Chris Romeo (34:47)**: Slightly different wording.

**Matt Coles (34:48)**: And then they have these cross-functional threats, which are actually also built into the layers separately. And you have a layer that's vertical.

**Chris Romeo (34:55)**: Yeah.

**Matt Coles (34:56)**: Bleh.

**Izar Tarandach (34:56)**: What? No, if you look at the step-by-step approach, right? It lists 6 steps. System decomposition.

**Chris Romeo (35:04)**: D, it's under D, using Maestro.

**Izar Tarandach (35:07)**: So system decomposition, layer-specific threat modeling, cross-layer threat identification, risk assessment, mitigation planning, implementation and monitoring.

**Matt Coles (35:20)**: Sure.

**Izar Tarandach (35:20)**: So I'm sorry.

**Matt Coles (35:23)**: What are we working on? What could go wrong? What are we going to do about it? And did we do a good enough job?

**Izar Tarandach (35:31)**: No, no, wait, wait, wait, wait, wait, wait. Let's be fair. Let's be fair. That is the root of everything, right? So kudos to Adam. He found the seed. But what I'm trying to point at here is that if you collapse layer-specific threat modeling with cross-layer threat identification, which for the love, for the life of me, I can't see why they have to be separate, right?

**Matt Coles (35:57)**: You shouldn't, because if you're going to do the layer cake, you're going to have communication between layers. Otherwise, why?

**Izar Tarandach (36:04)**: Anyway, I mean, it could be a matter of scoping. Let's look at this thing before we look at that thing. And that makes it easier to like focus on something at a time rather than just go crazy as sometimes we do.

**Chris Romeo (36:17)**: Yeah, I mean, I can actually justify the separation of the two from just trying to make it more simple to understand. If you're, if you got a room full of people who are newer to threat modeling, instead of just saying, throwing everything out on the table, okay, we're going to start with layer 1 foundation models. Here's the 5 things. Is there anything here? Let's go to layer 2, layer 3. Okay, now let's talk about the cross-layer.

**Izar Tarandach (36:40)**: Okay, so let's try that for a second.

**Matt Coles (36:41)**: Sure.

**Izar Tarandach (36:42)**: We are a team developing an agent. Okay. Just developing an agent.

**Matt Coles (36:48)**: Like I said, we use— is that layer— is that layer 5 or layer 7?

**Izar Tarandach (36:52)**: Yes. So we are just developing an agent here. We walk into the room. Everybody's in there. Okay, people, we are going to talk foundational models. Uh, we use ChatGPT, Claude, Gemini, Barkle. I don't know. Okay. No problems with foundational models.

**Chris Romeo (37:13)**: Move on.

**Izar Tarandach (37:14)**: Data operations. We buy it from, I don't know who. We just hold everything in the same place that we hold everything else. Okay, let's look at this one. There must be something in there. And then we get to layer 6, security and compliance, and everybody looks around and says, what have we been doing for the past 5 layers here? And because that's the vertical one, right?

**Matt Coles (37:34)**: Actually, I have a more concerning part to Chris's point earlier. If you're building an agent, that stops at layer 3, which is agent frameworks. And actually the agent itself, the agent itself is actually not part of these layers.

**Izar Tarandach (37:48)**: But wait, I'm sorry. It doesn't. It doesn't because it gets revisited in agents in layer 7, agent ecosystem.

**Matt Coles (37:56)**: Right. So is an agent of, is a, is a singular agent part of an ecosystem?

**Izar Tarandach (38:00)**: Yes.

**Chris Romeo (38:04)**: Sure.

**Izar Tarandach (38:04)**: Because it believes in an environment. It interacts with stuff.

**Matt Coles (38:10)**: Okay, so what you're saying is, if you have an agent, all 7 layers apply.

**Izar Tarandach (38:19)**: No, what I'm saying is, depending on how you're building that agent, depending on the choices that you are making or the choices that are given to you, that are impinged upon you. Some of the layers will make absolutely no sense to you because you can't do anything at that layer. And to tell me that you can solve lower level problems at a higher level, I, I, I, I will say no. Defense in depth, I mean, you need all the ducks aligned. You can't magically solve lower stuff up there.

**Matt Coles (38:51)**: Point of order. You may not be able to do anything about it other than make a decision about whether you want to do something about it.

**Izar Tarandach (38:58)**: Right? Yes.

**Matt Coles (39:00)**: No, that's understandable, right? If you have, if you're building on top of a foundation model and you identify that there's a chance of, if you, if you think about a foundation model and you say, oh, that foundation model has, is known to have data poisoning, I can make a decision to change models. I can't do anything about the data poisoning. Other than transfer the risk or change the risk.

**Izar Tarandach (39:27)**: That's right. But I'm going to be a bit cheeky here and offer that if you are doing your threat modeling of which model you should be using, at the moment that you are deciding to threat model your agent, you are very late in the game because this should really be different threat modeling sessions.

**Matt Coles (39:49)**: There's ideal. And then there's, there's actual in-practice reality.

**Izar Tarandach (39:54)**: And, and, and my point is the actual reprise in, in practice, we are pointing people to spend cycles thinking about things that, fine, we know that these problems exist. We can look at a catalog and say, these problems exist. And here's how each one of the foundational models says that they approach them. But at the level of I'm writing an agent. A, most probably you don't have the expertise to address those things. And B, most probably you're going to say the choice was made for me. There's nothing I can do.

**Matt Coles (40:29)**: But those are still decisions you can make. I mean, you can, you may not be able to address the threats that come out of it. You may not even know what all the threats are, although you can see the symptoms of it and you can make, I'm going to, I'll use the phrase, I didn't know it's the wrong phrase. An informed risk decision to do something about it.

**Izar Tarandach (40:48)**: A very smart person that I value very much said once that there is the practical and there is the theoretical. In theory, right? In practice, most people will look at that and say, again, I can't do anything about it. Why am I spending cycles on this?

**Chris Romeo (41:07)**: Sure.

**Izar Tarandach (41:08)**: But when you—

**Matt Coles (41:10)**: Go ahead. No, you finish your thought.

**Izar Tarandach (41:11)**: No, no, no, you go. No, no, no, you go.

**Matt Coles (41:14)**: But that in and of itself is a problem. Why am I spending cycles on something I can't change? Because you have the ability, you have to think about it in order to make something, an informed decision.

**Izar Tarandach (41:26)**: When I know how to answer that, my therapist is out of a job. So I can't say.

**Matt Coles (41:33)**: What I mean is, if you can't do anything about, if you Even though you can't do something— well, sorry. You can do something about it, but your choices are limited. Ignoring the problem because your choices are limited is a problem.

**Izar Tarandach (41:48)**: But that's the thing. You're not ignoring. You are looking at a— it's like any other, like, third-party component, right? It's bringing its threat modeling baggage with it.

**Matt Coles (41:59)**: That's right. And that may be opaque to you or not, right?

**Izar Tarandach (42:02)**: Yeah, yeah, yeah. Some of them, you can come and say, okay, I can do something with my defense in depth in here, put some bumps in the road. Some of them, you're going to say, hey, you know what, the decision was made. This is a base component of my design. I'm going to assume the risk. And both are valid. And both include looking at the problem and saying the problem exists. And there is very little I can do about it. My whole point here is that we are in a time in this technology where lumping things together for people that can't do anything about the stuff, and what they can do is very limited, is counterproductive at this time. If you are going to propose a framework to deal with agentic whatevers, and I'm just saying whatevers because I forgot the work, the word. It's not a whatevers. It's not a disrespect for whatever. It's just that I forgot the word. But if you're going to do a framework about agents, do the framework about agents. Create the framework in a way that people can actually respond to threats to agents, right? Don't try to boil the ocean. And go back to training models and whatever, because the mass, the 99% of people out there writing agents are not training models. There's enough complexity in here to address in a simple way each real layer of the complexity in ways that are actionable to that layer. And the spaghetti in my head says, you are juggling everything together. You, you, you are dropping a cognitive load on top of me of things that I am going to feel completely unempowered to, to solve. How exactly did you help me?

**Chris Romeo (44:03)**: Interesting. Hard-hitting.

**Matt Coles (44:05)**: That's deep. I just need some deep learning for that.

**Izar Tarandach (44:09)**: I, I, I blame it on Monster. I blame it on Monster.

**Chris Romeo (44:12)**: All right, well, we're, we're just about out of time here, so. I mean, when I, I, after diving into Maestro here, kind of kicking it around a little bit, maybe a little bit too much, I mean, I think there's still quite a bit of value in this, in this research and this work that's gone on. I think we're, we're, we're kind of looking at it for, we're trying to optimize it for enterprise use to some degree by saying it, it's needs to be refactored some to make it simpler and help people, personas understand how to use it. But all in all, I think there's some good stuff in here. And there's good threats that if this thing just gets repurposed as a threat library, I like that idea, Yasar. If it's a threat library, it's one of the best ones out there having to do with AI.

**Izar Tarandach (45:01)**: Look, I would say it's definitely a very, very worthy addition to the tools that we have. And from reading it, I already learned a bunch, right? And he won't let me lie about it.

**Matt Coles (45:18)**: I was waiting for the voice of approval over there.

**Izar Tarandach (45:21)**: Right? This is a, yeah, it's a Doug and a, how do you say?

**Matt Coles (45:32)**: Yeah, that.

**Izar Tarandach (45:32)**: There you go. Okay, he said everything. But basically, just from reading it, I learned a bunch. I learned stuff that I did not take into consideration before. So definitely a lot of value in there, a lot of great work in there. Goes together with a lot of great work that's been done in the OWASP GenAI project. Amazing stuff. I really think that, and actually just saying it in my head, It jumps, hey, if it bothers you so much, why don't you go and do it? Who knows? There is a level of translation that needs to happen here into turning this actionable to developers. And the three of us, our whole career, we are about making things actionable for developers. So that's probably what bothers us here so much. I know that that's what bothered me. But definitely a lot of good stuff in here. Yeah, I'm going to make use of it.

**Chris Romeo (46:31)**: Nice. All right. Well, Matt, any final thoughts on Maestro?

**Matt Coles (46:37)**: I'm going to try and get a word in edgewise with the barking dog in the background.

**Izar Tarandach (46:41)**: I would have loved—

**Matt Coles (46:43)**: It's not going to happen, but I wonder if—

**Chris Romeo (46:47)**: Folks, just so you know, Matt has no dogs. This is totally done in post-production just to try and—

**Matt Coles (46:52)**: She's just nice on board.

**Chris Romeo (46:53)**: To make him seem more human.

**Matt Coles (46:56)**: Yes, yes, that's what it is. Uh, you know, I was gonna drop a bombshell. I would really love if, uh, I would really have loved if this layer approach— that's them, and hopefully my voice is going to come through as opposed to— yeah, we hear it.

**Chris Romeo (47:12)**: Okay, we got you.

**Matt Coles (47:13)**: Uh, the, um, I would have loved if the layer approach was something we did wide I wish we could have used the layered approach. It's something we could apply to SAS things. Yeah, yeah, yeah, yeah. All right. I'm not going to do anything here.

**Chris Romeo (47:35)**: The layered approach. Matt wishes we could have used the layered approach for something.

**Matt Coles (47:40)**: For systems. For systems.

**Izar Tarandach (47:42)**: Hamburger.

**Matt Coles (47:44)**: Hamburger.

**Izar Tarandach (47:45)**: Hamburger.

**Matt Coles (47:45)**: The layered approach. The layered approach in my, I, as I read through it, the layered approach probably makes a lot of sense if we look at infrastructure processing data, right? As layers.

**Chris Romeo (47:56)**: Yeah.

**Matt Coles (47:57)**: In a more generalized way. And then take the threat library or the threat categories separately or the patterns separately. But altogether, um, I'll go back to Maestro is a is a reinvention of the wheel when the wheel was perfectly— was almost perfectly fine.

**Chris Romeo (48:18)**: Yeah.

**Matt Coles (48:20)**: How you use the wheel might, might have been the important bits.

**Chris Romeo (48:24)**: Yeah. And the wheel's had some improvements over time too. It has not made out of stone anymore or anything.

**Matt Coles (48:30)**: And now they don't even require air.

**Chris Romeo (48:32)**: True. Run forever. All right, folks. Well, thanks for tuning in to the Security Table. And hearing us discuss Maestro once again. We're not haters. We're just security people for a long time. So it just makes us how we are.

**Matt Coles (48:50)**: Cynical. Skeptic.

**Izar Tarandach (48:54)**: What? I don't believe you.
