RT Avi Chawla (@_avichawla)After MCP, A2A, & AG-UI, there's another Agent protocol (open-source).ACP (Agent Communication Protocol) is a standardized, RESTful interface for Agents to discover and coordinate with other Agents, regardless of their framework (CrewAI, LangChain, etc.).Here's how it works:- Build your Agents and host them on ACP servers.- The ACP server will receive requests from the ACP Client and forward them to the Agent.- ACP Client itself can be an Agent to intelligently route requests to t ...

Agent Communication Protocol (ACP) Overview - ACP (Agent Communication Protocol) is introduced as a new open-source Agent protocol [1] - The protocol facilitates Agent discovery and coordination, irrespective of their underlying framework (e g CrewAI, LangChain) [1] - ACP utilizes a standardized, RESTful interface [1] Resource and Contact Information - Avi Chawla (@_avichawla) shares tutorials and insights on DS, ML, LLMs, and RAGs [1] - A link is provided for further details on how ACP works (https://t co/q6xFvQKYgw) [1]

GitHub repo: https://t.co/FnehWkOqGD.Get a free visual guidebook to learn MCPs from scratch (with 11 projects): https://t.co/rQnuFGWGQd. ...

Agent Communication Protocols - ACP (Agent Communication Protocol) is presented as a new open-source protocol for agent communication, offering a standardized RESTful interface [1] - The protocol facilitates agent discovery and coordination across different frameworks like CrewAI and LangChain [1] - An ACP Client can function as an intelligent router for requests to agents, similar to MCP Client [1] ACP vs A2A - ACP is designed for local-first, low-latency communication, while A2A is optimized for web-native, cross-vendor interoperability [3] - ACP utilizes a RESTful interface for easier integration, whereas A2A supports more flexible interactions [3] - ACP is suited for controlled, edge, or team-specific environments, while A2A excels in broader cloud-based collaboration [3] Development and Deployment - The industry is encouraged to build ACP-compliant agents using frameworks like CrewAI and Smolagents [3] - Agents can be chained in sequential and hierarchical workflows, managed by ACP clients [3] - Agents can be imported to a public registry for easy discovery [3] - DeeplearningAI offers a course on building and hosting agents on ACP servers [2]

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs. https://t.co/AxraWu19RNAvi Chawla (@_avichawla):Postman's AI-readiness Playbook is one of the most important documents you can read today as a developer!We are headed into an era where every website must be "Agent-ready".- Agents will make purchases, not humans.- Agents will find the best options, not humans.- Agents https://t.co/ePqILv3UZX ...

Download the Playbook here: https://t.co/WQvpKyl4CR ...

AI Readiness & API Transformation - Every website must be "Agent-ready" in the coming era [1] - APIs need to be transformed into reliable, AI-ready tools [2] - Postman's 90-day AI readiness playbook details how to turn APIs into reliable, AI-ready tools [2] Key Components for AI-Ready APIs - Predictable structures are essential for AI agents [3] - Machine-readable metadata is crucial for AI understanding [3] - Standardized behavior is necessary for seamless AI interaction [3] Postman Playbook Highlights - Automatic documentation can be achieved by standardizing API format, Postman's Spec Hub automatically generates and validates API docs for both humans and AI agents without any manual work [2] - Validated specs can be turned into hosted, function-style endpoints, letting AI agents invoke APIs like native commands [3] Impact of AI Agents - Agents will make purchases, not humans [3] - Agents will find the best options, not humans [3] - Agents will fill out job applications, not humans [3]

LLM Application Evaluation - Deepeval enables component-level evaluation and tracing of LLM applications, addressing the need to identify issues within retrievers, tool calls, or the LLM itself [1] - The "@observe" decorator allows tracing of individual LLM components like tools, retrievers, and generators [2] - Metrics can be attached to each component for detailed analysis [2] - Deepeval provides a visual breakdown of component performance [2] Open Source and Data Control - Deepeval is a 100% open-source tool with over 8.5 thousand stars [2] - Users can self-host Deepeval to maintain control over their data [2] Ease of Use - Implementing Deepeval requires only 3 lines of code [1] - No refactoring of existing code is needed [1]

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs.Avi Chawla (@_avichawla):A Python decorator is all you need to trace LLM apps (open-source).Most LLM evals treat the app like an end-to-end black box.But LLM apps need component-level evals and tracing since the issue can be anywhere inside the box, like the retriever, tool call, or the LLM itself. https://t.co/dWXyJb3DNs ...

GitHub repo: https://t.co/LfM6AdsO74(don't forget to star it ⭐️) ...