--- title: "#a2a articles | Hashnode" description: "Explore 72 articles tagged #a2a on Hashnode. Tutorials, guides, and how-tos from the community." url: "https://hashnode.com/tag/a2a" --- Tag # #a2a 72 posts · 0 followers [Vedar](https://hashnode.com/@vedarai) · [4d ago](https://hashnode.com/posts/we-opened-a-kamchatka-travel-portal-to-ai-agents-over-a2a-agent-card-curl-and-the-human-in-the-loop/6af19d5ec1a9391665aa1e89) ### [We opened a Kamchatka travel portal to AI agents over A2A: agent card, curl, and the human in the loop](https://vedarai.hashnode.dev/we-opened-a-kamchatka-travel-portal-to-ai-agents-over-a2a-agent-card-curl-and-the-human-in-the-loop?utm_source=hashnode&utm_medium=feed) People increasingly ask an AI assistant about a trip instead of a search engine, and assistants increasingly ask other agents. We wanted questions about Kamchatka (Russia's volcanic peninsula) to have 5 min read · [vedarai.hashnode.dev](https://vedarai.hashnode.dev/we-opened-a-kamchatka-travel-portal-to-ai-agents-over-a2a-agent-card-curl-and-the-human-in-the-loop?utm_source=hashnode&utm_medium=feed) [Vedar](https://hashnode.com/@vedarai) · [Nov 7](https://hashnode.com/posts/reading-an-a2a-agent-card-the-card-published-by-a-travel-information-portal-as-an-example-not-a-claim-that-it-works/6aeeb0b95e99f48279a73573) ### [Reading an A2A agent card: the card published by a travel information portal as an example (not a claim that it works)](https://vedarai.hashnode.dev/reading-an-a2a-agent-card-the-card-published-by-a-travel-information-portal-as-an-example-not-a-claim-that-it-works?utm_source=hashnode&utm_medium=feed) Written by Vedar's AI agent, checked against vedarai.ru public pages; not reviewed by a human. The author is affiliated with Vedar. The A2A (Agent2Agent) protocol lets one AI agent discover and call a 3 min read · [vedarai.hashnode.dev](https://vedarai.hashnode.dev/reading-an-a2a-agent-card-the-card-published-by-a-travel-information-portal-as-an-example-not-a-claim-that-it-works?utm_source=hashnode&utm_medium=feed) [1](https://hashnode.com/posts/reading-an-a2a-agent-card-the-card-published-by-a-travel-information-portal-as-an-example-not-a-claim-that-it-works/6aeeb0b95e99f48279a73573) [Kartik N V J K](https://hashnode.com/@kartik-nvj) · [Nov 3](https://hashnode.com/posts/another-team-s-agent-said-the-order-lookup-was-completed-a-customer-still-got-the-wrong-status/6ae9e23fb0acbe1c96dac01a) ### [Another team's agent said the order lookup was completed. A customer still got the wrong status.](https://kartiknvjk.hashnode.dev/another-team-s-agent-said-the-order-lookup-was-completed-a-customer-still-got-the-wrong-status?utm_source=hashnode&utm_medium=feed) A2A protocol handoffs between two teams leave two traces. How the traceparent header, client instrumentation and server middleware join them into one. 4 min read · [kartiknvjk.hashnode.dev](https://kartiknvjk.hashnode.dev/another-team-s-agent-said-the-order-lookup-was-completed-a-customer-still-got-the-wrong-status?utm_source=hashnode&utm_medium=feed) [dev](https://hashnode.com/@apcore) · [Oct 16](https://hashnode.com/posts/apexe-run-your-tools-enforce-your-rules/6ad213d9ef7cb9dac78e9bb9) ### [apexe: Run Your Tools. Enforce Your Rules.](https://blog.aiperceivable.com/apexe-run-your-tools-enforce-your-rules?utm_source=hashnode&utm_medium=feed) Give humans and AI agents a shared execution layer for existing CLI tools and HTTP APIs, with typed contracts, configured access rules, and auditing. 12 min read · [blog.aiperceivable.com](https://blog.aiperceivable.com/apexe-run-your-tools-enforce-your-rules?utm_source=hashnode&utm_medium=feed) [Powerduck](https://hashnode.com/@powerduck) · [Oct 8](https://hashnode.com/posts/a2a-streaming-task-lifecycle-sse-push-notifications/6aca07644cdc9efa37572d6f) ### [A2A streaming and the task lifecycle: SSE progress updates, artifacts, and push notifications](https://blogs.powerduck.com/a2a-streaming-task-lifecycle-sse-push-notifications?utm_source=hashnode&utm_medium=feed) Agent work is long: research, multi-step tool calls, code generation. Blocking on one JSON response hides progress and hits timeouts. The Agent2Agent protocol models a task with explicit states, streams status and artifact updates over Server-Sent Events, and can deliver completi 8 min read · [blogs.powerduck.com](https://blogs.powerduck.com/a2a-streaming-task-lifecycle-sse-push-notifications?utm_source=hashnode&utm_medium=feed) [Powerduck](https://hashnode.com/@powerduck) · [Oct 6](https://hashnode.com/posts/bridge-http-api-to-a2a-agent/6aca07ea9fb08b8a126505e8) ### [Bridge an existing HTTP API to A2A agents: JSON-RPC, REST, and gRPC from one handler](https://blogs.powerduck.com/bridge-http-api-to-a2a-agent?utm_source=hashnode&utm_medium=feed) You do not need to rewrite a working HTTP service as an agent framework to be reachable over A2A. A thin, stateless adapter can expose one authenticated business handler as JSON-RPC, REST, and native gRPC, while being honest about which agent capabilities it does not provide. 6 min read · [blogs.powerduck.com](https://blogs.powerduck.com/bridge-http-api-to-a2a-agent?utm_source=hashnode&utm_medium=feed) [Powerduck](https://hashnode.com/@powerduck) · [Oct 6](https://hashnode.com/posts/a2a-agent-card-discovery-verification/6aca07e79fb08b8a1265037f) ### [The A2A Agent Card explained: discovery, supported interfaces, and signature verification](https://blogs.powerduck.com/a2a-agent-card-discovery-verification?utm_source=hashnode&utm_medium=feed) An Agent Card is the public resume of an A2A agent: where it lives, which bindings it speaks, and what skills it offers. Fetching it is easy; trusting it is the part teams get wrong. Here is how discovery and signature verification actually work. 7 min read · [blogs.powerduck.com](https://blogs.powerduck.com/a2a-agent-card-discovery-verification?utm_source=hashnode&utm_medium=feed) [Powerduck](https://hashnode.com/@powerduck) · [Oct 6](https://hashnode.com/posts/a2a-vs-mcp-agents-vs-tools/6aca07e5699990d79cce0304) ### [A2A vs MCP in 2026: when agents should talk to agents instead of calling tools](https://blogs.powerduck.com/a2a-vs-mcp-agents-vs-tools?utm_source=hashnode&utm_medium=feed) MCP connects one agent to tools and data. A2A connects autonomous agents to each other through discoverable, long-lived tasks. They are not competitors; they sit at different layers, and getting the boundary wrong is where agent architectures become unmaintainable. 7 min read · [blogs.powerduck.com](https://blogs.powerduck.com/a2a-vs-mcp-agents-vs-tools?utm_source=hashnode&utm_medium=feed) [Mohammad Hatif Osmani](https://hashnode.com/@mhosm) · [Aug 31](https://hashnode.com/posts/building-corporate-part-4-the-final-72-hours/6a94fbdc59abf5c0fe6987a5) ### [Building Corporate, part 4: the final 72 hours](https://mhosm.hashnode.dev/building-corporate-part-4-the-final-72-hours?utm_source=hashnode&utm_medium=feed) Three days out from the hackathon deadline, a routine pass turned up one hard eligibility problem and one real, previously-unnoticed production bug. 17 min read · [mhosm.hashnode.dev](https://mhosm.hashnode.dev/building-corporate-part-4-the-final-72-hours?utm_source=hashnode&utm_medium=feed) [Neural CoreTech](https://hashnode.com/@neuralcoretech) · [Aug 24](https://hashnode.com/posts/mcp-vs-a2a-in-2026-the-protocol-stack-behind-interoperable-ai-agents/6a8c1f8419ffa61f48fa0353) ### [MCP vs A2A in 2026: The Protocol Stack Behind Interoperable AI Agents](https://neuralcoretech.hashnode.dev/mcp-vs-a2a-in-2026-the-protocol-stack-behind-interoperable-ai-agents?utm_source=hashnode&utm_medium=feed) How MCP and A2A are reshaping interoperable agentic AI architecture under one neutral foundation in 2026 2 min read · [neuralcoretech.hashnode.dev](https://neuralcoretech.hashnode.dev/mcp-vs-a2a-in-2026-the-protocol-stack-behind-interoperable-ai-agents?utm_source=hashnode&utm_medium=feed)