I've been using the product now for about 2 weeks and it has been great. Only now have I run into my first real issue. I copied and pasted some text into AI context flow with hopes of going back to the text on a later date (today) to copy the text and paste it elsewhere only to find that I can't view the raw information ever again. Only the agent can view the raw information and tell me it's...
Founder Team
Hira_PluralityNetwork
May 15, 2026
A: We are currently working on improving the memory feature. The following features will be part of our next release:
1. View/download raw data 2. Edit the stored memory 3. Check versions of the individual memory items and restore older versions (git like version history)
These will be available both via MCP and inside the Memory Studio UI.
While you seem to be providing up to Opus 4.5, we are already at Opus 4.7, with another one around the corner. I know that, as technology progresses and new models are being released, it's hard to stay on top of things. However, some competitors already have Opus 4.7 and other models. I'm wondering how quickly you plan to add these models—specifically Opus 4.7—and whether there is a reason it...
Founder Team
Hira_PluralityNetwork
May 11, 2026
A: We have an integration with AWS Bedrock on the back in the EU region. We keep adding the newer models as they get rolled out.
Regarding images, videos and audios to be ingested as context, this is on the roadmap and will be added as we continue developing the product. I cannot say an exact ETA as the next few sprints are fully planned out but definitely within 6 months.
I am looking for a vector database to provide context to an LLM as part of a coaching app for our members. As you develop your OCL, will this become possible. My understanding is that right now, AI Context Flow is a private or personal storage only..
Founder Team
Hira_PluralityNetwork
Edited May 11, 2026
A: For business cases, we will create a specific package in the future.
But for now, you can use your app email to create an MCP connection with Access Tokens and then use that for your app's context.
Do you have the option to connect via MCP in your LLM? what's your setup?
Yes I do, for the mockup we are using Pickaxe.co ..
When you say "for business cases, you will create a specific package".. do you mean custom solutions or there is a new model that is generic but designed to support customer facing applications.
And of course, will this be available to LTD customers... This would make your app an essential for us, so fingers crossed.
I mean that if you end up with an application that needs more credits than what our current tiers offer and you keep hitting into limits, we will design specific token usage based packages for people building applications using our MCP as the memory/context layer.
But tier 3 should be a good starting point for you if you intend to go down this path.
Q: Straico integration
May I confirm if there will integration with straico?
Share AI Context Flow
Q: Are users locked out of the raw memory data?
I've been using the product now for about 2 weeks and it has been great. Only now have I run into my first real issue. I copied and pasted some text into AI context flow with hopes of going back to the text on a later date (today) to copy the text and paste it elsewhere only to find that I can't view the raw information ever again. Only the agent can view the raw information and tell me it's...
Hira_PluralityNetwork
May 15, 2026A: We are currently working on improving the memory feature. The following features will be part of our next release:
1. View/download raw data
2. Edit the stored memory
3. Check versions of the individual memory items and restore older versions (git like version history)
These will be available both via MCP and inside the Memory Studio UI.
Share AI Context Flow
Q: Viewing Request and Roadmap
Hey Support Team,
How about is your improvements PDF Viewing Request and Viewing for full details are saved in the platform? And how can users assess the roadmap?
Thank you.
Hira_PluralityNetwork
May 14, 2026A: Hi there,
We are currently working on the full viewing of individual memory contents and editing. These will go live in the next release.
Regarding the product roadmap, it's written in one of our earlier founder posts.
These are the two big milestones upcoming in next 6 months.
1. Team Plans: Context sharing within members of a team
2. Context marketplace: Create your buckets with...
Share AI Context Flow
Q: Opus 4.7, etc
While you seem to be providing up to Opus 4.5, we are already at Opus 4.7, with another one around the corner. I know that, as technology progresses and new models are being released, it's hard to stay on top of things. However, some competitors already have Opus 4.7 and other models. I'm wondering how quickly you plan to add these models—specifically Opus 4.7—and whether there is a reason it...
Hira_PluralityNetwork
May 11, 2026A: We have an integration with AWS Bedrock on the back in the EU region. We keep adding the newer models as they get rolled out.
Regarding images, videos and audios to be ingested as context, this is on the roadmap and will be added as we continue developing the product. I cannot say an exact ETA as the next few sprints are fully planned out but definitely within 6 months.
Share AI Context Flow
Verified purchaser
Do you have a public roadmap or anything? We can kind of see what maybe the timeline is.
Q: OCL - Public Facing App Development
I am looking for a vector database to provide context to an LLM as part of a coaching app for our members. As you develop your OCL, will this become possible. My understanding is that right now, AI Context Flow is a private or personal storage only..
Hira_PluralityNetwork
Edited May 11, 2026A: For business cases, we will create a specific package in the future.
But for now, you can use your app email to create an MCP connection with Access Tokens and then use that for your app's context.
Do you have the option to connect via MCP in your LLM? what's your setup?
Share AI Context Flow
Verified purchaser
Yes I do, for the mockup we are using Pickaxe.co ..
When you say "for business cases, you will create a specific package".. do you mean custom solutions or there is a new model that is generic but designed to support customer facing applications.
And of course, will this be available to LTD customers... This would make your app an essential for us, so fingers crossed.
I mean that if you end up with an application that needs more credits than what our current tiers offer and you keep hitting into limits, we will design specific token usage based packages for people building applications using our MCP as the memory/context layer.
But tier 3 should be a good starting point for you if you intend to go down this path.