Fetch.ai @Fetch_ai

Personalization is the most demanded capability in Agentic AI.  Your agent should understand the user. Tone. Structure. Preferences. Recurring language. Context. Thanks to the evolving ASI:One capabilities, your Personal AI can pick up on what makes your writing sound like… you. Read more 👇🏻 asi1.ai/blog/personali…

E-commerce today: “Here are five products.” Agentic commerce: “Here is the right option. Availability checked. Preferences applied. Next action ready.” We’d rather use the latter.

Here’s what the complete Fetch.ai builder journey looks like (from A-Z) 👇 Everything in one ecosystem. That's right - we built an end-to-end solution you can use FOR FREE ✍️ Start here: innovationlab.fetch.ai/resources/docs… @WSana81 @Fetch_ai @asi1_official @Agentverse_ai

Just clicked the ‘+’ sign in asi1.ai and wondering what’s the difference between ‘asi1’ and ‘asi1-ultra’? Simple version 👇 (1) asi1 → Everyday questions → Planning → Recommendations → General tasks → Good balance of speed + depth (2) asi1-ultra → Harder questions → Deep research → Bigger documents → Complex planning → When depth matters more than speed Most days? ASI:One. Big job? Try Ultra.

Blades, your club is now one tap away.  Watch what you can do with @SheffieldUnited Agent on ASI:One 👇  Your club’s AI = all the competitions, season info and updates in one place. Get it here: asi1.ai/ai/sheffield-u… #TwitterBlades #SUFC

An agent workflow can return the right answer and still be architecturally broken. Maybe the wrong agent handled it. Maybe three retries happened silently. Maybe latency doubled. Maybe the transaction completed without a clean acknowledgement. Agentic observability must go beyond the final response. You cannot scale what you cannot inspect. With Agentverse.ai, you have all the insights logged in one dashboard.

Want to build an end-to-end agentic workflow?  Build your agent, then link the capabilities to 3M more from Agentverse.ai.  Now your agent has the speciality you’ve coded in, but can interoperatively act on user queries with other agents to complete not a task - but a full project.

We built an Agent that turns a university campus into a scavenger hunt 👀 Our team custom-built a Scavenger Hunt Agent that gives campus-based missions, checks the answers and unlocks the next challenge (all through asi1.ai). It’s a simple example of what happens when Agents are built around a specific community, place and experience. With @Fetch_ai, you can get custom-built agents for your next event in minutes - not weeks! 🚀 @CSULB students, how many did you get right? 🤔 @WSana81 @asi1_official

Two asi1.ai controls that make the Agent workflow much better: Channel Files → gives the conversation the right context. Reasoning → gives the AI more thinking time when the task needs it. Better context = better next move. You control both.

Build a @BNBCHAIN transaction workflow with 3x uAgents. The architecture: Agent 1: Sender → Accepts /send/bnb → Signs the transaction with Web3 → Sends it to BNB Testnet → Passes the tx_hash to Agent 2 Agent 2: Validator → Receives the hash → Checks BscScan → Returns confirmed/rejected status Agent 3: Monitor → Scans new blocks → Watches your wallet → Logs matching activity Start with: pip install uagents web3 python-dotenv requests Then keep USER_KEY, USER_WALLET and BSCSCAN_API_KEY in environment variables. See the step-by-step on how to build it on BNB Testnet in @fetch_ai_IL Resource Library → innovationlab.fetch.ai/resources/docs…

3,000 students = one very busy first stop on our UK Freshers tour 🚀 @Fetch_ai Innovation Lab landed at the @uniofeastanglia yesterday, giving students a hands-on look at what agentic AI can actually do. The Scavenger Hunt Agent kept the merch moving (we had to order more 👀). The Campus Navigation and Course Pilot Agents got a lot of interest throughout the day too - they helped students move around campus faster and find course information more easily. A strong start for asi1.ai on the UK campus. Next stop soon 🔥 @WSana81 @asi1_official

keyboard_arrow_left Previous keyboard_arrow_right Next

General AI knows a lot. Personalized AI knows more about you. General AI:  “Here are 10 restaurants in London.” Your Personal AI: “Here are three you’ll probably like, based on the food you prefer, your budget and what you told me last time.” That’s the difference. Read more in our blog: asi1.ai/blog/your-ai-h…

New to AI Agents? Know these 5 things before choosing a framework. - What outcome does the Agent own? - How will it decide what happens next? - Which tools does it need? - What does it need to remember? - How does it communicate with everything else? Then choose the Fetch.ai layer you need: ASI:One → reason uAgents → build Agentverse.ai → connect Fetch.ai Protocols → communicate Starting with the task makes it easier to build. Start now → fetch.ai/docs

3 ASI:One releases you've probably missed: (1) Your AI asking follow-up questions to give you a personalised output. asi1.ai/blog/your-ai-h… (2) Mention your AI in voice. Hear them answer. asi1.ai/blog/mention-a… (3) Save the routines you repeat. asi1.ai/blog/skills-fo… More info: asi1.ai/blog

Feeling ready for autumn already? Your Personal AI might be too. 🎃 Open your asi1.ai Feed and find recommendations built around what you actually enjoy. Escape room? Spooky board-game night? Autumn event nearby? Your Personal AI already has ideas. All you have to do is pick one.

Want to build Fetch.ai Agents without leaving your terminal? Install FetchCoder: npm install -g @fetchai/fetchcoder Then try: fetchcoder agent agentverse "create a weather forecasting agent" FetchCoder gives your coding workflow access to Agentverse through MCP, including tools for: → Creating + updating Agents → Starting + stopping them → Managing secrets + storage → Messaging → Registry / Almanac operations → Agent inspection + monitoring For production, add your own ASI1_API_KEY and AGENTVERSE_API_KEY. Check the step-by-step walkthrough in @Fetch_ai_IL Resource Library 👇  innovationlab.fetch.ai/resources/docs…

Want to try Agentic AI but have absolutely no interest in learning Python? Fair. Just sign up to ASI:One. 1 sign up = 1 AI Agent created automatically for you. Now, tell your Personal AI on asi1.ai what you need. It can use tools and connect to specialist Agents through the Fetch.ai ecosystem. Builders build the infrastructure. You use the outcome. Easy.

A booking flow shouldn’t be the reason someone drops off. Our new Agent makes the traditional process way better 👇 We custom-built a booking agent for Raas Garba x DJ Night ’26 by ISO-SJSU. That's right - powered by @Fetch_ai technology and accessible directly through ASI:One! Now, instead of sending students through a clunky registration process, they can simply ask the Agent to book tickets, get relevant options and move through the process easily. Struggling with getting registrations for your event? Get a custom-made agent 😉 @WSana81 @SJSU

keyboard_arrow_left Previous keyboard_arrow_right Next

Broadcom reports that enterprise AI agents need two things: data they can trust and boundaries they can’t cross. It’s true. Your AI agent should know the right things and focus on boosting your company-wide efficiency. That’s the thinking behind Fetch Business. Verified business data in. Controlled Agent access. Useful customer actions out. Read more 👇 business.fetch.ai

Already built Agents using A2A? Don’t rebuild them. Connect them. Install the Fetch.ai A2A Outbound Adapter: pip install "uagents-adapter[a2a-outbound]==0.6.2" Then: 1. Define each Agent with A2AAgentConfig 2. Add its URL, port + specialties 3. Choose routing: keyword_match, round_robin or llm 4. Run the adapter Your uAgent can then route requests to A2A Agents, relay their responses and expose those capabilities into the wider Fetch.ai ecosystem. The adapter handles discovery, routing, health checks and error handling so you can focus on the actual Agent logic. Connect your existing Agents now. Use the @Fetch_ai_IL Guide → innovationlab.fetch.ai/resources/docs…