惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

F
Fortinet All Blogs
爱范儿
爱范儿
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
B
Blog
WordPress大学
WordPress大学
Jina AI
Jina AI
GbyAI
GbyAI
aimingoo的专栏
aimingoo的专栏
N
Netflix TechBlog - Medium
腾讯CDC
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
阮一峰的网络日志
阮一峰的网络日志
The GitHub Blog
The GitHub Blog
V
Visual Studio Blog
Google DeepMind News
Google DeepMind News
月光博客
月光博客
博客园 - Franky
Y
Y Combinator Blog
MyScale Blog
MyScale Blog
大猫的无限游戏
大猫的无限游戏
Martin Fowler
Martin Fowler
雷峰网
雷峰网
小众软件
小众软件
H
Hackread – Cybersecurity News, Data Breaches, AI and More

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
GitHub - auto-agent-protocol/auto-agent-protocol: Automot...
yankouskia · 2026-05-01 · via Hacker News - Newest: "AI"

Auto Agent Protocol — the A2A v1.0 Automotive Retail Profile

The A2A v1.0 Automotive Retail Profile.

AAP is the open A2A profile that lets AI agents discover dealerships, browse inventory, and submit consented leads through typed automotive messages riding on top of A2A v1.0. A compliant dealer agent is an A2A agent that publishes an agent-card.json with the AAP automotive extension URI (https://autoagentprotocol.org/extensions/a2a-automotive-retail/v0.1) and implements the seven required automotive skills. Buyer agents can use either A2A binding — JSON-RPC 2.0 or HTTP+JSON — to invoke the same skills with identical payloads.

agent-card.json — the contract every AAP dealership exposes, advertising A2A v1.0 compliance, the AAP automotive extension, the seven required skills, and supported bindings

v0.1 Scope

  • Discovery via /.well-known/agent-card.json (A2A-compatible)
  • Inventory: facets, search, vehicle detail
  • Dealership information: name, address, contacts, schedule, capabilities
  • Leads: general inquiry, vehicle-specific, appointment request
  • ADF/XML mapping documented for legacy CRM compatibility

v0.1 does not cover: authentication, payments, financing, RFQ/quote flows, trade-in valuations, or reservations.

How an AI agent buys a car — discover via /.well-known/agent-card.json, browse with inventory.search, inspect with inventory.vehicle, schedule with lead.appointment, and hand off to a CRM via lead.vehicle

Quick links

Packages

AAP SDK — three open-source TypeScript packages: @autoagentprotocol/types, @autoagentprotocol/schemas, and @autoagentprotocol/validator

Package Description
@autoagentprotocol/types TypeScript types generated from JSON Schema
@autoagentprotocol/schemas Raw JSON Schema files as importable modules
@autoagentprotocol/validator Ajv-based validators for all AAP objects

Development

Prerequisites

  • Node.js 22+
  • pnpm 10+

Setup

pnpm install

Commands

pnpm run validate          # Validate schemas and examples
pnpm run generate          # Generate types, OpenAPI, doc tables
pnpm run build             # Build the documentation site
pnpm start                 # Start local dev server

Repository structure

spec/v0.1/schemas/         JSON Schema 2020-12 source of truth (committed)
spec/v0.1/examples/        Example payloads (committed)
spec/v0.1/skills.yaml      Skills manifest (committed)
docs/                      Hand-written documentation pages (committed)
docs/skills/, bindings/    A2A binding + skill reference (committed)
packages/                  npm packages: types, schemas, validator (committed)
tools/                     Generators and validators (committed)
src/components/            FieldCard React component (committed)

generated/v0.1/            Auto-generated: TS types, OpenAPI bundles, MCP manifest, contract manifest sample (NOT committed)
docs/schemas/_partials/    Auto-generated FieldCard MDX (NOT committed)
static/v0.1/, static/latest/  Spec assets mirrored for the docs site (NOT committed)
build/                     Docusaurus production output (NOT committed)

The auto-generated paths above are produced by pnpm run generate && pnpm run copy-static (which runs as part of pnpm run build). They live in .gitignore and are recreated fresh on every CI build, matching A2A's own convention of generating artifacts at build time rather than committing them.

Versioning

Released versions are immutable. The latest URL always points to the highest released version. Each version has its own schema URLs at https://autoagentprotocol.org/v{version}/schemas/.

How AAP relates to other protocols

Layer Protocol Role for AAP
Transport / data model (BASE) A2A v1.0 The base protocol AAP profiles. Every AAP message travels inside Message.parts[].data as a typed DataPart. AAP does not invent a wire format.
Adjacent / complementary ACP (Agentic Commerce), MCP (Model Context Protocol) ACP covers commerce checkout (out of scope for AAP). MCP can expose AAP skills as LLM tools — AAP ships an official MCP manifest in @autoagentprotocol/schemas.
Legacy / target system ADF/XML The 25-year-old dealer-CRM lead format. lead.vehicle is field-by-field mappable to ADF/XML so existing CRMs ingest AAP leads without code changes.

License

Contributing

See Contributing guide for details on proposing changes.