Creator's Box
@ctrsbox.bsky.social
June 3, 2025 at 10:00 AM
To try it out, head over to: genspark.ai/
And make sure to include the word “Download” in your prompt
And make sure to include the word “Download” in your prompt
June 2, 2025 at 1:30 PM
To try it out, head over to: genspark.ai/
And make sure to include the word “Download” in your prompt
And make sure to include the word “Download” in your prompt
5. “Download this Instagram video: [URLs]”
June 2, 2025 at 1:30 PM
5. “Download this Instagram video: [URLs]”
4. Download 20 European landmark building pictures from copyright-free image websites to Al Drive.”
June 2, 2025 at 1:30 PM
4. Download 20 European landmark building pictures from copyright-free image websites to Al Drive.”
3. Ask it to “Download multiple TikTok videos”
June 2, 2025 at 1:30 PM
3. Ask it to “Download multiple TikTok videos”
2. "Download 20 dog images from copyright-free websites”
June 2, 2025 at 1:30 PM
2. "Download 20 dog images from copyright-free websites”
1. Download all papers from a LinkedIn post
June 2, 2025 at 1:30 PM
1. Download all papers from a LinkedIn post
Handoff
The handoff feature enables agents to call other agents to complete tasks or hand over a conversation mid-action.
Handoff enables complex multi-agent orchestration to support even more advanced and capable agentic systems.
docs.mistral.ai/agents/agen...
The handoff feature enables agents to call other agents to complete tasks or hand over a conversation mid-action.
Handoff enables complex multi-agent orchestration to support even more advanced and capable agentic systems.
docs.mistral.ai/agents/agen...
May 28, 2025 at 10:12 AM
Handoff
The handoff feature enables agents to call other agents to complete tasks or hand over a conversation mid-action.
Handoff enables complex multi-agent orchestration to support even more advanced and capable agentic systems.
docs.mistral.ai/agents/agen...
The handoff feature enables agents to call other agents to complete tasks or hand over a conversation mid-action.
Handoff enables complex multi-agent orchestration to support even more advanced and capable agentic systems.
docs.mistral.ai/agents/agen...
Function Calling
Besides the built-in tool and MCP servers, it also supports function calling. This enables devs to build their own functions and call them during agent conversations.
Besides the built-in tool and MCP servers, it also supports function calling. This enables devs to build their own functions and call them during agent conversations.
May 28, 2025 at 10:12 AM
Function Calling
Besides the built-in tool and MCP servers, it also supports function calling. This enables devs to build their own functions and call them during agent conversations.
Besides the built-in tool and MCP servers, it also supports function calling. This enables devs to build their own functions and call them during agent conversations.
MCP Support
You can also connect agents with MCP servers. Supports local MCP servers, remote MCP servers, and remote MCP servers with Auth.
You can also connect agents with MCP servers. Supports local MCP servers, remote MCP servers, and remote MCP servers with Auth.
May 28, 2025 at 10:12 AM
MCP Support
You can also connect agents with MCP servers. Supports local MCP servers, remote MCP servers, and remote MCP servers with Auth.
You can also connect agents with MCP servers. Supports local MCP servers, remote MCP servers, and remote MCP servers with Auth.
Agent Connectors
You can connect tools like web search and code execution to the agents.
Other built-in tools include image generation and a document library (accessing documents from Mistral Cloud) for building agentic RAG systems.
You can connect tools like web search and code execution to the agents.
Other built-in tools include image generation and a document library (accessing documents from Mistral Cloud) for building agentic RAG systems.
May 28, 2025 at 10:12 AM
Agent Connectors
You can connect tools like web search and code execution to the agents.
Other built-in tools include image generation and a document library (accessing documents from Mistral Cloud) for building agentic RAG systems.
You can connect tools like web search and code execution to the agents.
Other built-in tools include image generation and a document library (accessing documents from Mistral Cloud) for building agentic RAG systems.
Then you can start a conversation or continue the conversation.
You need a conversation id to continue the conversation.
It also supports output streaming.
You need a conversation id to continue the conversation.
It also supports output streaming.
May 28, 2025 at 10:12 AM
Then you can start a conversation or continue the conversation.
You need a conversation id to continue the conversation.
It also supports output streaming.
You need a conversation id to continue the conversation.
It also supports output streaming.
Agents Basics
It's super simple to create an agent by giving it a description, name, instructions, and tools.
It's super simple to create an agent by giving it a description, name, instructions, and tools.
May 28, 2025 at 10:12 AM
Agents Basics
It's super simple to create an agent by giving it a description, name, instructions, and tools.
It's super simple to create an agent by giving it a description, name, instructions, and tools.
There is a nice documentation for this release.
You can see below the things that are supported.
Persistent state across conversations, image generation, handoff capabilities, structured outputs, document understanding, citations, and more.
You can see below the things that are supported.
Persistent state across conversations, image generation, handoff capabilities, structured outputs, document understanding, citations, and more.
May 28, 2025 at 10:12 AM
There is a nice documentation for this release.
You can see below the things that are supported.
Persistent state across conversations, image generation, handoff capabilities, structured outputs, document understanding, citations, and more.
You can see below the things that are supported.
Persistent state across conversations, image generation, handoff capabilities, structured outputs, document understanding, citations, and more.
5. Vy AI Agent sees and uses your computer just like you do.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
May 27, 2025 at 10:00 AM
5. Vy AI Agent sees and uses your computer just like you do.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
4. Scout by Scrapybara is an AI agent with its own virtual computer. It can do deep research, coding, data analysis, and more. Start with a prompt and any files.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
May 27, 2025 at 10:00 AM
4. Scout by Scrapybara is an AI agent with its own virtual computer. It can do deep research, coding, data analysis, and more. Start with a prompt and any files.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
2. Agent Neo by Flowith is the first agent with unlimited context, steps, and output.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
May 27, 2025 at 10:00 AM
2. Agent Neo by Flowith is the first agent with unlimited context, steps, and output.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
3. Works directly inside GitHub
No new dashboard. No context switching.
→ Jules connects to your GitHub
→ Creates a new branch
→ Pushes PRs with commit messages + explanations
→ You review + approve
It's like onboarding a new engineer who doesn’t sleep.
No new dashboard. No context switching.
→ Jules connects to your GitHub
→ Creates a new branch
→ Pushes PRs with commit messages + explanations
→ You review + approve
It's like onboarding a new engineer who doesn’t sleep.
May 26, 2025 at 1:00 PM
3. Works directly inside GitHub
No new dashboard. No context switching.
→ Jules connects to your GitHub
→ Creates a new branch
→ Pushes PRs with commit messages + explanations
→ You review + approve
It's like onboarding a new engineer who doesn’t sleep.
No new dashboard. No context switching.
→ Jules connects to your GitHub
→ Creates a new branch
→ Pushes PRs with commit messages + explanations
→ You review + approve
It's like onboarding a new engineer who doesn’t sleep.
2. Jules runs in its own secure VM
Your repo gets cloned into a private Google Cloud VM.
Jules gets full context no guessing, no hallucinations.
It then drafts a plan → shares its reasoning → edits your code.
All changes are visible as diffs before merge.
Your repo gets cloned into a private Google Cloud VM.
Jules gets full context no guessing, no hallucinations.
It then drafts a plan → shares its reasoning → edits your code.
All changes are visible as diffs before merge.
May 26, 2025 at 1:00 PM
2. Jules runs in its own secure VM
Your repo gets cloned into a private Google Cloud VM.
Jules gets full context no guessing, no hallucinations.
It then drafts a plan → shares its reasoning → edits your code.
All changes are visible as diffs before merge.
Your repo gets cloned into a private Google Cloud VM.
Jules gets full context no guessing, no hallucinations.
It then drafts a plan → shares its reasoning → edits your code.
All changes are visible as diffs before merge.
1. What is Jules?
Jules isn’t a co-pilot.
It’s a full agentic developer that reads your repo, understands your intent, and makes multi-file edits on its own.
• Write tests
• Fix bugs
• Upgrade dependencies
• Build new features
• Audio changelogs (!)
All async. All in the cloud.
Jules isn’t a co-pilot.
It’s a full agentic developer that reads your repo, understands your intent, and makes multi-file edits on its own.
• Write tests
• Fix bugs
• Upgrade dependencies
• Build new features
• Audio changelogs (!)
All async. All in the cloud.
May 26, 2025 at 1:00 PM
1. What is Jules?
Jules isn’t a co-pilot.
It’s a full agentic developer that reads your repo, understands your intent, and makes multi-file edits on its own.
• Write tests
• Fix bugs
• Upgrade dependencies
• Build new features
• Audio changelogs (!)
All async. All in the cloud.
Jules isn’t a co-pilot.
It’s a full agentic developer that reads your repo, understands your intent, and makes multi-file edits on its own.
• Write tests
• Fix bugs
• Upgrade dependencies
• Build new features
• Audio changelogs (!)
All async. All in the cloud.
5. Vy AI Agent sees and uses your computer just like you do.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
May 24, 2025 at 12:00 PM
5. Vy AI Agent sees and uses your computer just like you do.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
Vy works across all applications you're signed into, remembers your information, and can handle professional software even if you don't know how to use it.
Available now for MacOS.
4. Scout by Scrapybara is an AI agent with its own virtual computer. It can do deep research, coding, data analysis, and more. Start with a prompt and any files.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
May 24, 2025 at 12:00 PM
4. Scout by Scrapybara is an AI agent with its own virtual computer. It can do deep research, coding, data analysis, and more. Start with a prompt and any files.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
Scout browses the web, runs terminal commands, edits code, and creates files on its virtual computer.
2. Agent Neo by Flowith is the first agent with unlimited context, steps, and output.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
May 24, 2025 at 12:00 PM
2. Agent Neo by Flowith is the first agent with unlimited context, steps, and output.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
Give it literally any task - it spins up multiple AI agents on-the-fly, working in parallel to complete it.
Can execute 1000+ reasoning steps with 10M+ context.
7. Anthropic rolled out 4 major features to their API to build powerful agentic apps:
→ Code execution to run Python code directly in the API call
→ MCP connector to connect to remote MCP servers from the API
→ Files API to upload documents once
→ Extended context caching
→ Code execution to run Python code directly in the API call
→ MCP connector to connect to remote MCP servers from the API
→ Files API to upload documents once
→ Extended context caching
May 23, 2025 at 10:00 AM
7. Anthropic rolled out 4 major features to their API to build powerful agentic apps:
→ Code execution to run Python code directly in the API call
→ MCP connector to connect to remote MCP servers from the API
→ Files API to upload documents once
→ Extended context caching
→ Code execution to run Python code directly in the API call
→ MCP connector to connect to remote MCP servers from the API
→ Files API to upload documents once
→ Extended context caching
6. Vibe code complete, cross-platform mobile apps using AI and React Native, powered by Claude 4.
Rork lets you build one-shot apps that use GPT-4o API.
Another amazing part of this release - the video is created completely using Google Veo 3 model.
Rork lets you build one-shot apps that use GPT-4o API.
Another amazing part of this release - the video is created completely using Google Veo 3 model.
May 23, 2025 at 10:00 AM
6. Vibe code complete, cross-platform mobile apps using AI and React Native, powered by Claude 4.
Rork lets you build one-shot apps that use GPT-4o API.
Another amazing part of this release - the video is created completely using Google Veo 3 model.
Rork lets you build one-shot apps that use GPT-4o API.
Another amazing part of this release - the video is created completely using Google Veo 3 model.