In a recent Claude Code session I tried using the Google Docs, Drive, and Sheets MCP and was honestly surprised at how limited it felt. It was hard to get anything meaningful done because it just did not expose enough capability to be useful in practice. In hindsight, that frustration was probably a good thing. I ended up skipping MCP entirely and using the LaTeX API plus its plugin ecosystem, and the result was far beyond anything I could have realistically produced through Docs anyway.
I have seen a similar pattern with Canva’s MCP. I pay for Pro, but the one feature that would actually make MCP useful, Auto Fill, is gated behind an enterprise plan. So the surface is there, but the real power is locked away.
I get that this is still the wild west for MCP, and I agree with the OP’s general take. But right now there is a big gap between "integration exists" and "integration is actually useful." Personally, I am more excited about where something like WebMCP could go, where the default assumption is full capability rather than a restricted subset.
The result is less that I want to go to Figma directly and more that I just want to skip it entirely. So, assuming the power of these aggregator agents keeps growing, the onus is on these tools to create useful integrations or get subsumed by a model capability or another tool with a better integration. It sounds like your experience is similar - you bypassed the tools with bad integrations instead of going to them directly.
You have to be very very careful with this stuff. These SaaS companies have tons of paying customers giving them thousands of dollars a month. If customers mess up with an officially supported MCP and delete their assets or break implementations or DDOS their own site it’d be nightmare for sales / support.
It makes sense they very slowly transitioned from read-only to limited write. You have to carefully beta test. Both figuring out the guardrails and finding usecases where it actually works well. The only way to do that (properly) is a slow drip release cycle.
I attended a design conference last week where Figma has been basically delegated as a design library tool, and that was it. They'd use it as a source of truth for components such as buttons, colors, typography, etc, but the actual design work that was being done was done through Claude Code. Multiple designers who had this stack said that they preferred it as their designs were now closer to what the end user would experience (i.e. code). One person actually eschewed Figma completely and used Storybook as the source of UI truth. I think that Figmas moat is a lot smaller than people think and that within a year or 2 there's going to be some very solid competitors out there.
SaaS products are headed to where the UI isn't the undifferentiated factor. People have been so busy building a UI that works for everyone so it works for no one. The real value is the data and the workflow they provide. Make the data accessible to agents (MCP, OpenClaw skills, etc0.
Completely agree. Every SaaS tool will come with an MCP or an API to leverage composability. We can unlock useful functionalities from Claude Code and other aggregators (terminology from the post) to be able to compose different MCP's from different SaaS. One can imagine composing the results from a google search and using it in for a Figma design attempt, as a simple example.
This is an obvious direction that the industry is heading to. But what are the implications of this? I think the differentiating factor of having a good UI will reduce - productivity apps and SaaS will no longer have their aesthetic UI as a moat. I'm not sure whether this will tank their stocks or increase the valuation but what I'm sure of is that the productivity and usage will increase.
The combinatorial utility of different MCPs / APIs inside an agent is an interesting angle. Figma can technically plug into the same MCPs and use all the same models, but if the software design process doesn’t start in Figma anymore, it does not matter. The value will accrue to the point of integration (the agent).
Re: the value of good UI/UX
I think short-term, the value of good interfaces will actually increase - if anyone can easily build out the same product in 10 different ways, the best designed one that people actually want to use will likely be the choice. But that’s holding constant lots of things like distribution, type of SaaS, its place in the transaction stack, etc. So either guess would make a lot of assumptions.
It also appears (so far at least) that these models really struggle with front-end design. Something like /frontend-design skill is good but only gets you so far. It still requires a ton of steering to get it to a sensical place. So for now, whoever can steer it is still valuable. But I’m sure more and more of that will get codified and internalized by the model and the harness. So the design steering will become more and more abstract.
Long-term, we’re likely moving towards dynamically generated interfaces. Claude is already doing it with diagrams and charts in the chat. This opens up so many fascinating questions. What happens when UI doesn’t have to be one-size-fits-all, where each person may get their GUI generated with their preferences and context in mind? What happens to the design process when your UI doesn’t have to scale to a ton of user types and use cases? Will we even be designing UIs or something else entirely? Will Jakob’s Law still apply or will our individual GUIs diverge so much that I won’t be able to navigate your smartphone if I pick it up?
First, mcp, like cli, like api, is a kind of (user facing) interface, just not graphical. It still needs to be designed, just by different people with differing skills.
Second, textual interface can only go so far in terms of information ingestion. Trying to describe a complex relationship between entities can be extremely difficult with text. However, a good graphical interface will make complex information easier to digest.
So in my opinion, the moat will emphasize organization which knows how to plan good a custom experience, whether graphical or textual, and less where tables and forms are the main business
Previously, a lot of SaaS’s valuation was dependent on it being a ‘platform’ where their customers ‘do almost everything only on their platform’, keeping them within that SaaS’s ecosystem.
By making the tools compostable, the valuation from this ‘keeping within’ angle will slowly disappear, but maybe it can be replaced by increase usage as a source of valuation.
I think that also explains why a bunch of companies are now all racing to build the same thing - the everything-in-one-place universal context store with their own agents on top of it. Linear, Notion, Salesforce, etc. Because the alternative is a much worse business to be in.
I noticed that too, but I wonder if what they are offering is just a proprietary formatting for context store, or is there something more operationally complex than that.
I mean, this was the web 2.0 dream. And then everyone realized that giving people an easy way out of your platform wasn't good for business. And all of the APIs dried up. Tremendously disappointing.
Actually... their MCP reflects more than just supporting AI. It demonstrates the future of closed, walled garden MCPs.
While Figma advertises that their official MCP "can now write directly to your Figma files", in reality it is restricted to create and read (as in CRUD), but not update nor delete. Currently, there is only one option to edit/update using the Figma MCP and it requires going through a third-party service with its own subscription and tiny token allocations.
Meanwhile, several developers have figured out how to use Figma's plug-in system to work-around these limitations, for a more robust CRUD MCP solution.
I don’t think so. Your product / tool has to be extremely specialized and deeply necessary to pull of a walled garden MCP. The more likely alternatives are that someone else comes along with a better (open) integration, the model internalizes your tool’s capabilities, or the AI labs themselves build 1st party competitors.
It's so excellent that the walls are coming tumbling down.
Walls and moats are just not viable in the way it used to be: the market tightly demands agency, and it's not something your company can provide solo. The point is that users want to intermix their experiences.
Figma's been super super punished by the market (down from $80 ish ipo in August to $20-$30 since Jan). The faith that AI is somehow is going to build such a mass adopted product that works so relatively well; seems weird. Figma really made strong up front choices to make such strong bones for their work: it feels rare to see companies go for such deliberately chosen high technology web, and it feels like these choices (wasm, multiplayer, etc) have been so core to building such a great business.
Definitely don’t believe AI is going to “one-shot” Figma. But Figma was built for a world where product design is a stand-alone step in a series of discrete steps with hand-offs between them. It was a place where design started and ended, for a while at least. If people don’t see it as a starting point anymore, it becomes a very different business & product.
In a recent Claude Code session I tried using the Google Docs, Drive, and Sheets MCP and was honestly surprised at how limited it felt. It was hard to get anything meaningful done because it just did not expose enough capability to be useful in practice. In hindsight, that frustration was probably a good thing. I ended up skipping MCP entirely and using the LaTeX API plus its plugin ecosystem, and the result was far beyond anything I could have realistically produced through Docs anyway.
I have seen a similar pattern with Canva’s MCP. I pay for Pro, but the one feature that would actually make MCP useful, Auto Fill, is gated behind an enterprise plan. So the surface is there, but the real power is locked away.
I get that this is still the wild west for MCP, and I agree with the OP’s general take. But right now there is a big gap between "integration exists" and "integration is actually useful." Personally, I am more excited about where something like WebMCP could go, where the default assumption is full capability rather than a restricted subset.
I agree. I also tried Figma’s MCP tools (wrote about it here - https://metedata.substack.com/p/metedata-digest-003-where-do...) and found it very underwhelming.
The result is less that I want to go to Figma directly and more that I just want to skip it entirely. So, assuming the power of these aggregator agents keeps growing, the onus is on these tools to create useful integrations or get subsumed by a model capability or another tool with a better integration. It sounds like your experience is similar - you bypassed the tools with bad integrations instead of going to them directly.
You have to be very very careful with this stuff. These SaaS companies have tons of paying customers giving them thousands of dollars a month. If customers mess up with an officially supported MCP and delete their assets or break implementations or DDOS their own site it’d be nightmare for sales / support.
It makes sense they very slowly transitioned from read-only to limited write. You have to carefully beta test. Both figuring out the guardrails and finding usecases where it actually works well. The only way to do that (properly) is a slow drip release cycle.
I attended a design conference last week where Figma has been basically delegated as a design library tool, and that was it. They'd use it as a source of truth for components such as buttons, colors, typography, etc, but the actual design work that was being done was done through Claude Code. Multiple designers who had this stack said that they preferred it as their designs were now closer to what the end user would experience (i.e. code). One person actually eschewed Figma completely and used Storybook as the source of UI truth. I think that Figmas moat is a lot smaller than people think and that within a year or 2 there's going to be some very solid competitors out there.
SaaS products are headed to where the UI isn't the undifferentiated factor. People have been so busy building a UI that works for everyone so it works for no one. The real value is the data and the workflow they provide. Make the data accessible to agents (MCP, OpenClaw skills, etc0.
People who do that will do well.
Completely agree. Every SaaS tool will come with an MCP or an API to leverage composability. We can unlock useful functionalities from Claude Code and other aggregators (terminology from the post) to be able to compose different MCP's from different SaaS. One can imagine composing the results from a google search and using it in for a Figma design attempt, as a simple example.
This is an obvious direction that the industry is heading to. But what are the implications of this? I think the differentiating factor of having a good UI will reduce - productivity apps and SaaS will no longer have their aesthetic UI as a moat. I'm not sure whether this will tank their stocks or increase the valuation but what I'm sure of is that the productivity and usage will increase.
The combinatorial utility of different MCPs / APIs inside an agent is an interesting angle. Figma can technically plug into the same MCPs and use all the same models, but if the software design process doesn’t start in Figma anymore, it does not matter. The value will accrue to the point of integration (the agent).
Re: the value of good UI/UX
I think short-term, the value of good interfaces will actually increase - if anyone can easily build out the same product in 10 different ways, the best designed one that people actually want to use will likely be the choice. But that’s holding constant lots of things like distribution, type of SaaS, its place in the transaction stack, etc. So either guess would make a lot of assumptions.
It also appears (so far at least) that these models really struggle with front-end design. Something like /frontend-design skill is good but only gets you so far. It still requires a ton of steering to get it to a sensical place. So for now, whoever can steer it is still valuable. But I’m sure more and more of that will get codified and internalized by the model and the harness. So the design steering will become more and more abstract.
Long-term, we’re likely moving towards dynamically generated interfaces. Claude is already doing it with diagrams and charts in the chat. This opens up so many fascinating questions. What happens when UI doesn’t have to be one-size-fits-all, where each person may get their GUI generated with their preferences and context in mind? What happens to the design process when your UI doesn’t have to scale to a ton of user types and use cases? Will we even be designing UIs or something else entirely? Will Jakob’s Law still apply or will our individual GUIs diverge so much that I won’t be able to navigate your smartphone if I pick it up?
>Long-term, we’re likely moving towards dynamically generated interfaces.
Right on point. Prediction: we will have a new protocol like HTML but for LLMs so that SaaS can communicate on this language.
Several points -
First, mcp, like cli, like api, is a kind of (user facing) interface, just not graphical. It still needs to be designed, just by different people with differing skills.
Second, textual interface can only go so far in terms of information ingestion. Trying to describe a complex relationship between entities can be extremely difficult with text. However, a good graphical interface will make complex information easier to digest.
So in my opinion, the moat will emphasize organization which knows how to plan good a custom experience, whether graphical or textual, and less where tables and forms are the main business
Previously, a lot of SaaS’s valuation was dependent on it being a ‘platform’ where their customers ‘do almost everything only on their platform’, keeping them within that SaaS’s ecosystem.
By making the tools compostable, the valuation from this ‘keeping within’ angle will slowly disappear, but maybe it can be replaced by increase usage as a source of valuation.
I think that also explains why a bunch of companies are now all racing to build the same thing - the everything-in-one-place universal context store with their own agents on top of it. Linear, Notion, Salesforce, etc. Because the alternative is a much worse business to be in.
I noticed that too, but I wonder if what they are offering is just a proprietary formatting for context store, or is there something more operationally complex than that.
I mean, this was the web 2.0 dream. And then everyone realized that giving people an easy way out of your platform wasn't good for business. And all of the APIs dried up. Tremendously disappointing.
We'll see if this time, things end differently.
The potential productivity increase with composable MCP's is too high for walled gardens to still sustain
100%. Have been saying for a long time that AI needs data, context and specific capabilities.
Figma offered the capabilities but not the data or context.
Everyone's building "the everything app". The end game is likely an entire OS shell that is AI-first.
Actually... their MCP reflects more than just supporting AI. It demonstrates the future of closed, walled garden MCPs.
While Figma advertises that their official MCP "can now write directly to your Figma files", in reality it is restricted to create and read (as in CRUD), but not update nor delete. Currently, there is only one option to edit/update using the Figma MCP and it requires going through a third-party service with its own subscription and tiny token allocations.
Meanwhile, several developers have figured out how to use Figma's plug-in system to work-around these limitations, for a more robust CRUD MCP solution.
I don’t think so. Your product / tool has to be extremely specialized and deeply necessary to pull of a walled garden MCP. The more likely alternatives are that someone else comes along with a better (open) integration, the model internalizes your tool’s capabilities, or the AI labs themselves build 1st party competitors.
Inter-Op is back! (Limited to AI overlord access only, please don't make our business redundant)
It's so excellent that the walls are coming tumbling down.
Walls and moats are just not viable in the way it used to be: the market tightly demands agency, and it's not something your company can provide solo. The point is that users want to intermix their experiences.
Figma's been super super punished by the market (down from $80 ish ipo in August to $20-$30 since Jan). The faith that AI is somehow is going to build such a mass adopted product that works so relatively well; seems weird. Figma really made strong up front choices to make such strong bones for their work: it feels rare to see companies go for such deliberately chosen high technology web, and it feels like these choices (wasm, multiplayer, etc) have been so core to building such a great business.
Definitely don’t believe AI is going to “one-shot” Figma. But Figma was built for a world where product design is a stand-alone step in a series of discrete steps with hand-offs between them. It was a place where design started and ended, for a while at least. If people don’t see it as a starting point anymore, it becomes a very different business & product.