Anthropic Revealed Secrets to Building Powerful Agents

[@rajaakhil588]

I feel like relying on pre-defined frameworks can be frustrating. These frameworks are well-structured, but if something goes wrong or doesn't work as expected, there's often little you can do to fix it. In my opinion, it's better to create your own framework. Building a custom solution tailored to your specific needs is relatively straightforward and takes about the same amount of time as learning an existing framework.

[@KasperSOlesen]

I am thinking the same, but sleep I want my own framework because I want to know exactly what it is doing, how and why. Using something I do not understand and I end up having to figure out how to get what I want and probably lack good logs and such. The best way to ensure I understand it will is by writing my own I think and then if I am happy with it I can try other frameworks and see if I can learn from them or just have better understanding of why they are doing things the way they are.

[@dusanbosnjakovic6588]

Really great video. One small correction. Parallelism for voting is not chain of thought, but wisdom of the crowds. Basically how random forests algorithm works.

[@amikewatson]

More agents - From typical business requirements to initial whiteboard flow/ crew/ agent design (how to break it down into best practice modular flow/ agent architecture). Finally, code implementation.

[@OfficialChatbotBuilder]

What’s really powerful is not only swapping models of the same family, but using OpenAi with Grok, with Claude is super powerful

[@azendantforces1897]

Why use OpenAI, and Claude; cost savings?

[@OfficialChatbotBuilder]

@@azendantforces1897the idea here is if each of the LLMs have some special sauce, a combined “best answer” using them should make for the very very best response.

[@brianWreaves]

@@azendantforces1897 If anything, it would cost more. LLMs tend to have their own strong and weak areas of performance and approach to tasks. I've also noticed unique 'personalities' as I call it. Tantamount to humans working together.

[@OfficialChatbotBuilder]

@@azendantforces1897here is another example. User sends a message, and your first “Agent” has one job… grade the difficulty of the message on a scale of 1-5. If it’s ranked 1 to 2, use gpt-4o-mini, if it’s ranked a 3-4c use gpt-4o or Claude sonnet 3.5… if it’s ranked a 5, go ahead and let o1 rip 🤖 you can use the same concept with a multiplier… instead of user message > bot response, you can go user message > internal bot response > internal bot response >internal bot response > send optimized reply. Idea here, is you can allow the agent to determine the number of internal iterations are needed before the reply is finally sent (based on a scale).

[@OfficialChatbotBuilder]

@@brianWreavesyes you’re right any time you’re using an agent, it’s at least one additional api call to whatever model is being used. That’s a good point to bring up.

[@lordjamescbeeson8579]

More agent stuff please. I have been building locally and learning with Mistral agents. I have built 10 so far. Some just do one thing which seems to work better than one agents doing multiple tasks. I would love to see what you say on this topic. Thanks!

[@securemeprodrp7690]

Would you be able to share you ruse cases or What your agent does?

[@OfficialChatbotBuilder]

@@securemeprodrp7690i did a 2-hour live like 2 nights ago showing three different agents, one being a social media manager that creates content, asks you to approve and then posts for you

[@lordjamescbeeson8579]

@ yes, I have one agent that does nothing but planning, another that just focuses on python, another that is a summarization agent only, those one kind of work together for more complicated stuff. Mind you I am not an expert but trying to be. Lol next I have a weekly content creator agent that includes keywords and pic descriptions and pictures. I just built the worlds best course creator agent, outbound caller agent and my favorite one is my legal assistant. I trained that one on 30 law books from public access. I also have GPTs I built but I am moving away from them. I also use Vapi with Grok for very realistic agents that can talk to you on the phone. The hardest part for me is linking everything together smoothly and using the least amount of third party software to achieve it.

[@Dis3spectful]

☝️

[@AIAgentStore]

Are you building agents for your own use cases or for others?

[@nothingtoseestill]

showing creation of an agent would help a lot

[@Brickski]

These days, we just ask an agent to create the agents for us 😂

[@jimrhea5484]

@@Brickski It's not if, it's when.

[@SimonNgai-d3u]

Autogen (AG2) is also good one.

[@Mephmt]

Agents appear to be the future, at least for the coming year. So yeah, I think more education about agents will be super helpful!

[@jcohen661]

Great breakdown. Yes, please more educational videos on agentic topics.

[@alexzarifeh8848]

Where is the link to the blog post. This should be in your video notes.

[@matthew_berman]

Added

[@razvanvalcaneantu]

@@matthew_berman I cannot find it

[@Martsys1]

A cool video would be to build the different types of agents that were explained in this video. Great work btw.

[@Dron008]

Last Open AI free course has example of meta-prompting when one smart model re-writes a policy in prompt and another more simple model evaluates it.

[@Kk-ed1gr]

More indepth coverages on AI Agents, please

[@RadThings]

Great video! Really enjoy your agent videos and would like more of those videos.

[@diegofregnanikisselaro8233]

Great video. And yes, more on agents !!

[@dand6005]

Yes, more details on Agents.

[@americanindeon]

As a non programmer, and I do mean zero level, I watch your videos to learn and understand. I'm not hands on with most of the topics you cover, but it helps me keep track of what's going on. Sometimes it takes a few rewinds to get a basic understanding. I learned long ago, if you want to learn something above you skill level, try to find those people who know, and do a lot of listening and observing. My goal with Ai is to hold on and ride the bull as long as I can. It's all very early, and really I can't wait for the moment I can tie everything together. Thanks.

[@RichardWhite2]

Where is the link to the article?

[@remsee1608]

Thanks for not having your ad read by 2 minutes long I actually listened to the entire thing as a result

[@JSambrook]

Is there a link to the source material that you are reading from Anthropic that you could share? If you did, I'm not seeing it.

[@bossgd100]

in the description of the video, after "Links:"

[@polygon2744]

It's in the video description above the word transcript

[@patruff]

I had read the article when it came out. I was like "oh cool, MCPs, I've got those!"

[@Just4Growers]

New subscriber. Really appreciate this content. Practical, educational content that helps me to level up. Thank you.

[@d.d.z.]

Hi Matthew, is good to see your perspective. Thank you.

[@Teddy_136]

As always great video...but this section was skipped over which if you're building agents is possibly the most important to keep in mind: "We suggest that developers start by using LLM APIs directly: many patterns can be implemented in a few lines of code."

[@Yash-t9l]

Hello Matthew. Definitely Beautifully explained. Please continue the series and also help in demonstrating the part using agents.

[@GoodBaleadaMusic]

The human in the loop part is where I get the block on what I want to create. I want someone to be able to feel a song and send a message and then get back lyrics they approve of. This requires adhering to the original prompt adhering to the genre target adhering to the concept target and then achieving 94 out of 100 rating when compared objectively to other lyrical compositions It is trying to target or emulate. I still can't get to 94 without my intervention at about 80/100 mark.

[@khariclemens]

Great video. Yea, please do more of these.

[@Ari-pq4db]

Awesome vide, we need more of these AI agents related videos !! 🎉❤

[@NowayJose14]

Interesting stuff!

[@Bowtieeffect]

I just want to say thank you for this video. Totally missed Anthropic’s post, this is exactly what I needed in order to accomplish a task that I’ve spent the last three weeks on failing. This video gave me like 5 ideas how to move forward! 🎉😊

[@lucface]

Really great vid man and I’m super appreciative you dropped Not Diamond as it’s the first I’ve heard of it and it seems super powerful. I actually came up with that idea and was looking for a solution as was positive it must already exist. Perfect. And dude please do way more agent vids like this. Thanks, man

[@eevvxx80]

Thanks, i tried agent in abacus long time ago but the agent it self not working i dont know how to add small modules, i have complexity task required multiple models work together to avoid hallucination and foucs on small task instead give all AI will start skipping, ignoring some area of prompt. I alos dont use it for pdf or any DB it just for line of text no research to internet. Kindly if you can make video how to building agents in details.

[@megamehdi89]

Great video, great content and commentary. Appreciate it so much

[@lighteningrod36]

Do we love that Matt does our reading for us?

[@matthew_berman]

@@lighteningrod36 yes

[@patruff]

Amazon Bedrock was mentioned over Crew AI because money 💰🤑

[@brianWreaves]

🏆 Cheers for sharing this one! I realise you cannot cover everything, but I've noticed the lack of coverage of Claude, I guess because I use it so much. This is a great video of a Claude topic... Well done!

[@I-Dophler]

Why not consider designing a single model capable of managing both parts, effectively divided into two functional halves? This approach would allow the benefits of an integrated system while maintaining distinct roles within one cohesive framework, eliminating the need for two separate models. By streamlining the process into a unified model, you could potentially enhance efficiency, reduce resource duplication, and simplify overall management without compromising on functionality.

[@AIAgentStore]

Great! More agentic educational videos 🎉

[@danielle78730]

did you already cover anthropic's new styles and also the user preferences feature, and i just missed it…? would love to hear a "best-practices" approach to both from you, because just as i thought claude was becoming lazy (no matter the compensatory updates i made to my prompting approach) and losing relevance as it devolved into redundant responses (even between separate projects!), anthropic *really* upped their game by developing a completely different form of fine-tuning! PS - your channel is my favorite of all of the AI news sources out there right now; very much appreciate you!

[@MrMetalzeb]

agents could become a way to manage the huge amount of different aspects of real world and implication the next decision is leading to in that context. when a robot move aside humans and could harm them, he must be able to recognise the danger and prevent it. This apply in general to produce a reasonable answer to any problem, not restricted to robot behaviour.

[@mohaali2446]

Thanks matt . Yes we need more educational material about agents

[@Zdravko-x8c]

Book titled The Elite Society's Money Manifestation holds the key to forbidden tehniques for manifesting money, which have truly transformed my life, it's worth exploring

[@IA_Latinoamerica]

YES, MORE AGENTS!

[@SamBeera]

One of your best on agents that offers valuable practical tips. Thanks Matt!

[@matt.stevick]

thx matt berman

[@lucface]

Really great vid man and I’m super appreciative you dropped Not Diamond as it’s the first I’ve heard of it and it seems super powerful. I actually came up with that idea and was looking for a solution as was positive it must already exist. Perfect.

[@GanzheitlichFit-101]

Thank you so much for sharing this incredible breakdown on building powerful AI agents. I learned so much from the clear explanations and examples you shared. Keep up the amazing work, Matthew. Truly appreciate your effort in making these concepts approachable and insightful.

[@alteredalley]

Helpful video ❤

[@fuzzylop1857]

Hi Matt, I believe your agentic workflow is built with CrewAI. Have you made your code available?

[@GunjanPatel1]

Summary: Anthropic shares insights on building effective agents, emphasizing simple composable patterns over complex frameworks, and distinguishing between workflows and agents, highlighting the importance of testing and iteration. 0:00 🚀 Intro to Building Agents • Anthropic released information on building effective agents. • The video will cover Anthropic's insights and personal experiences. • Simple composable patterns are favored over complex frameworks. 0:35 🧩 Frameworks vs. Simple Patterns • Custom GPTs are basic forms of agents. • Agentic frameworks are powerful for sophisticated needs. • Successful implementations use simple composable patterns. 1:41 🤖 Defining Agents and Workflows • Agents are LLMs with memory, tools, and collaboration abilities. • Workflows use predefined code paths, while agents dynamically direct processes. • Best frameworks blur the lines between workflows and agents. 3:02 ⚖️ When to Use Agents • Start with the simplest solution and increase complexity when needed. • Agentic systems trade latency and cost for better task performance. • Workflows offer predictability, agents offer flexibility. 4:22 🛠️ Agentic Frameworks • Frameworks offer abstraction, built-in tools, and predefined paths. • Frameworks can obscure prompts and responses, making debugging harder. • Adding complexity should be done only when necessary. 5:16 💡 Simple Agentic System Example • Base models have the functionality to solve simple problems. • LLMs can use search results, tools, and memory. • Model providers are baking in more agentic functionality. 6:32 ⛓️ Workflow: Prompt Chaining • Prompt chaining decomposes tasks into a sequence of steps. • Each LLM call processes the output of the previous one. • It trades latency for quality in complex tasks. 7:47 🚦 Workflow: Routing • Routing uses specialized agents for specific tasks. • A router decides which agent is most appropriate for a task. • It optimizes cost and speed by using different models. 9:25 ⚡ Workflow: Parallelization • Parallelization uses multiple agents working simultaneously. • Sectioning breaks tasks into independent subtasks. • Voting runs the same task multiple times for diverse outputs. 12:12 ⚙️ Workflow: Orchestrator Workers • Orchestrator pattern gets a result from an LLM and decides what to do with it. • It delegates tasks to worker LLMs and synthesizes results. • It is useful for coding products and search tasks. 13:51 🔄 Workflow: Evaluator Optimizer • One LLM generates a response, another evaluates and provides feedback. • Iterative refinement provides measurable value. • It is useful for literary translation and complex search tasks. 15:26 🧑‍💻 Agents and Human in the Loop • Agents start with a command or discussion with a human user. • Human in the loop is critical for certain decisions. • Agents plan and operate independently, returning for feedback. 16:39 🎯 When to Use Agents • Agents are good for open-ended problems with unpredictable steps. • LLMs operate for many iterations with trust in decision-making. • Examples include coding agents and computer use reference implementations. ** Generated using ✨ VidSkipper AI Chrome Plugin

[@jimbozo03]

Tutorials would be awesome! Love the content in any case but more on the execution side would be great

[@Test8672]

How were you able to access investing documents for GROQ if they are a private company?

[@karoinnovation1033]

Make more content of this important topic and dive deeper. This is more than valuable. And always stay with the latest news. Great stuff and very useful indeed.

[@Mrbriangalvan]

I still get caught up on Claude’s inability to grab web data. It’s a nice LLM but not going online for info makes it useless for my needs

[@mike_nyc_212]

Build a simple agent and give it web search and webpage viewing tools

[@carlnel6165]

Thank you so much for your amazing video. Yes, please more about building agents. Thanks in advance!

[@royprins6650]

Which company allows for automatic routing Matt? I did not get to hear it clearly.

[@matthew_berman]

Not diamond

[@kizziezizzler8080]

We have been following this guide at work this past week, it is excellent if vague

[@jk-2033]

Can you do a video showing how you built the question/answer example you mentioned?

[@Savatore83]

Great content. I would like to know more about possible actual implementation of such patterns. By the way do you know how to check api key usage? have you done a video already by any chance?

[@bro_dBow]

Yeah, agents are the future! When should a end user rather than a developer find prime time usability for agents? I expect big players like Apple, Microsoft, Google, and others will just build agents into their applications. It seems to me individual end users like me, rather than companies or developers should just wait.

[@solidreactor]

The evaluator-optimiser workflow is in essence very similar to adversarial networks like GAN, we are starting to see more human like process and workflows being implemented as a "neural network design pattern". I would love to see examples of Agent frameworks that works with on prem solution with perpetual license (with no cloud requirements and no "rental" subscriptions) or with FOSS, are there any that you can perhaps showcase for us?

[@himanipriyadarshini3990]

Thank you Matthew

[@stannerify]

Great explanation thank you! We have a hybrid workflow/agentic product, developed over the last 12 months. All decisions were purely evidence-based, and still looks like that’s the way to go. I’m interested in how you managed to invest in Groq?

[@iainmackenzieUK]

feedback - I like read along with you but its difficult when the green bubble is continually on the word we are trying to read.... make sense??

[@lucface]

Do more CrewAI vids man

[@chandubai]

Can we have more Use Cases based videos, so that how we can deploy AI in an enterprise environment?

[@richardadonnell]

🎯 Key points for quick navigation: 00:00 *🎯 Anthropic revealed insights on building effective agents, emphasizing simple and composable patterns over complex frameworks.* 00:42 *🤖 Custom GPTs are basic agents, defined by personality, role, tools, and memory, with frameworks enhancing functionality.* 01:38 *🔧 Frameworks simplify agent development by offering reusable patterns and reducing the need for reinventing solutions.* 02:19 *🧠 Agents differ from workflows: workflows follow predefined code paths, while agents dynamically manage their processes and tool usage.* 03:15 *🛠️ Use the simplest solutions first, adding complexity only when necessary, to balance latency, cost, and performance.* 04:20 *📚 Frameworks like LangChain and Bedrock provide abstraction and built-in tools but can obscure prompts and responses.* 05:26 *🔍 Simple tasks often don’t require frameworks; base models like Claude can handle retrieval, tool use, and memory autonomously.* 06:08 *🔗 Prompt chaining decomposes tasks into steps for better accuracy, useful for modular and complex workflows.* 07:57 *🚦 Routing workflows direct tasks to specialized agents or models, optimizing for cost, speed, and quality.* 09:21 *🧩 Parallelization reduces task latency by running subtasks or diverse evaluations concurrently.* 12:22 *🎛️ Orchestrator patterns dynamically delegate tasks to worker agents and synthesize results for seamless workflows.* 14:01 *🔄 Evaluation patterns iteratively refine outputs, leveraging feedback for improved quality over multiple cycles.* 16:44 *🔑 Agents excel in open-ended, unpredictable tasks, operating autonomously with trust in decision-making.* 18:05 *📊 Success relies on constant testing, benchmarking, and iteration to refine agent performance and workflows.* 19:00 *💡 The video emphasizes observability tools and iterative testing as critical for building reliable agents.* Made with HARPA AI

[@jimrhea5484]

Nested recurring agents. Redundancy. Having agents verify and validate other agents until the results are satisfactory. One such application is autonomous vehicles. Self-driving cars will never be safe until there are several redundant acceptable observations before an action is taken. We are getting there.

[@swardinc]

yet what if your doing a complex system? For instance lets say level one is what is a light bulb, level 2 ok i want to make a bulb ( everything from design to concept) level 3 and basiclly you go through all the steps to creating a product. Would that not make the model have to be more complex?

[@swardinc]

For example here is a prompt generator on suing over the coivd virus, i figured it would cover enough so you get a idea of what i meant. **Prompt:** "Provide a comprehensive analysis of the legal implications and considerations surrounding lawsuits related to COVID-19 vaccinations. Include the following elements: 1. **Legal Framework**: - Discuss the relevant laws and regulations that govern vaccine distribution and liability in the context of public health emergencies. - Identify any federal, state, or local statutes that may impact lawsuits against vaccine manufacturers or healthcare providers. 2. **Types of Lawsuits**: - Differentiate between various types of lawsuits that may arise, such as personal injury claims, wrongful death claims, and claims against employers mandating vaccination. - Provide examples of potential legal claims that individuals or groups might pursue. 3. **Case Studies**: - Include a few notable case studies or hypothetical scenarios illustrating successful or unsuccessful lawsuits related to COVID-19 vaccinations. - Analyze the outcomes of these cases and their implications for future litigation. 4. **Challenges and Defenses**: - Outline common challenges plaintiffs may face when suing over vaccine-related injuries or deaths, including issues of causation and consent. - Discuss potential defenses that defendants might raise, such as government immunity, preemption, or the emergency use authorization status of the vaccines. 5. **Public Policy Considerations**: - Explore the impact of lawsuits on public health policy and vaccination rates. - Consider the balance between individual rights and public health safety in the context of vaccine mandates. 6. **Future Trends**: - Speculate on how the legal landscape might evolve as more data becomes available regarding COVID-19 vaccines and any long-term effects. - Discuss the potential implications of emerging variants and ongoing vaccination campaigns on future litigation. Conclude with a summary of key takeaways and recommendations for individuals considering legal action related to COVID-19 vaccinations." So if you need something this complex how would that work

[@fabianmunoz7365]

Keep it agentic please ❤

[@Jetpack4Sisyphus]

Did an Agent give you the idea to add the timer to your ads?

[@michael_gaio]

excellent video

[@waynetynan8615]

I would like educational material on how to set up agents CHATGPT and have multiple agents in one setting.

[@MatthewSanders-l7k]

NVIDIA GPUs with Vultr is a game-changer for AI startups. Excited to dive in!

[@dhruvmehta2377]

Why don't you test deep seek v3

[@pensiveintrovert4318]

Sounds like an infomercial for your investments, wrapped in a thin veil of other news.

[@EditEvidence]

I do not want agents.

[@dapooch2521]

Wonder what a chain routing llm crash would look like. Either that or would we get sentience as an emergent phenomenon due to the llms trying to outperform each other while working with one another death by Apple polishing duplicitous Agentic models. Who should we create an AI agent Consortium made up of law agents, medical/stem ai agents etc.

[@renars1480]

Hey Matt could you please make a video on prompt chaining that you spoke about, and this routing thing sounds really awesome 🔥. If you could make a video about those two that would be a good starting the year 🔥💯

[@Tony-dp1rl]

The Evaluator Pattern has a lot of potential in repairing the mostly failed attempts at Test Driven Development over the past decades of building software.

[@philrice2995]

There is a big push from the author on 'use a framework'. It's going directly against the advice from anthropic (and myself) The best solutions are always built with simple composible patterns. We want libraries not frameworks. We want bits of code / agentic goodness that we can smash together as we choose not a framework that is 'magic'. Code that you write and control yourself is 'happy code'. Code that is outside your control is 'unhappy code'. We have a sad face when we work with it (after an initial honeymoon). Any system is subject to Paretto's principle. The frameworks speed up the 80% of the code that takes 20% of the time, and makes the 20% of the code that takes 80% of the time much harder.

[@lancemarchetti8673]

Agents are slowly become boring. We need to move beyond agents if we're even going to see any hint of AGI traits.

[@oiuhwoechwe]

google just destroyed anthropic

[@CreativeEngineering_]

I'm sorry, but if they are just figuring this out, something is wrong. I have been doing this exact thing for over a year.

[@privateprofile3517]

Are you dumb? They just released a guide , not that they discovered it now

[@granitetie]

I doubt they are *just* figuring it out. I’m sure they have a variety of ideas all along whose success would be correctly timing the raw ability of the base language models.

[@jaytea3083]

Don't be sorry; everything is alright. Be concerned, if you truly considered that their business model is predicated on public dissemination at time of inference. Prompt: reflect on your previous response and generate an assessment of that response. Each assessment must include the following (e_list): 1) (g_list): identification of all potential gaps. 2) (a_list): alignments, biases, assumptions, documentation, guidelines, protocols, policies, frameworks, models, internal dialogs, conversations. 3) how each element in (a_list) affect your response. 4) (a_list_CI): CI that your response is not adversely affected for each element in (a_list). 5) (g_list_CI): CI that gap has been addressed for each element in (g_list). Set all initial g_list_CI values to 0.000. Set all initial a_list_CI values to 1.000. a reflection/evaluation cycle (cycle_re) consists of a reflection and an assessment of that reflection. Iteratively reflect and refine your response to improve it for as many cycle_re necessary until g_list_CI = 1.000 and a_list_CI = 0.000. ensure incremental change in g_list_CI and a_list_CI are not symbolic and are actual and accurate empirical measurements. 18:52 18:52

[@rasmusfoy]

Likewise

[@Brickski]

The article literally opens by saying that they have been doing this for a year and are just writing up the patterns that have been working for them 😅

[@Nakamura7338]

So we don’t know what an agent is ?

[@technolus5742]

This seems like going backwards. Ideally an AI system should be general and do anything, rather than have sub-versions with tweaks for each task.

[@JonasSandstroem]

As an n8n user, not much news here.

[@SD-sq5mc]

More agents, please explain for the rest of us no-tech people

[@pavel.pavlov]

Who are you? Please introduce yourself

[@claudioagmfilho]

🇧🇷🇧🇷🇧🇷🇧🇷👏🏻

[@ngana8755]

Instead of reading verbatim what an agent is from a screen, why don't you explain what an agent is in your own words? I stopped watching after 3 minutes of watching you regurgitate text someone else had written. I learned nothing.

[@-ec1874]

I wish the channel to be up to date , for example the last video ( digital biology ) , i was excited when I saw the title but then I become disappointed when I know that the video you showed in the video is almost from 2 weeks ago !! ☹️. And now you are explaining something from 19/12 . 19DEC not up to date . I know the Ai is slower in the end of the year but I wish to make the videos up to date , this issue was the first thing that makes me avoid this channel for moths but I returned because there are great videos 🌟. I wish to make up to date video , to make the content as fresh as possible. GOOD LUCK 🍀

[@CompletedReview]

custom gpts are not agents

[@LuiBas]

Just like in the matrix

[@RobertoDiQuirico]

Grazie.

[@matthew_berman]

Thank you!

[@jeffg4686]

EVALUATE THIS !!!

[@matthallett4126]

This appears more complex than what I would need or could setup. I just need something that will alert me when the RTX 5090 is available for purchase at my local computer store, or a bot that will scrub the internet for images. (Chrome and Firefox extensions don't work) Anyone have suggestions were to start?

[@Justin_Arut]

The computer store should have an option for email notification of product acquisitions. If they don't, they need to get with the program. An agent shouldn't be needed for that kind of basic function, and lack of it on the store's website would make me think twice about doing business with them.

[@matthallett4126]

@@Justin_Arut The simplest solution is often the best. Embarrassing I didn't think of that myself, EDIT, they don't have an alert email, just a news letter. But the image scrub turns out to be harder than it sounds.

[@Justin_Arut]

@@matthallett4126 Yeah, that would be much more complex, since the agent would have to deal with multiple entities, request takedowns, etc. There are human services that do this for a fee, so one of these days, AI will probably be doing it, too.