A2UI
Shares rank 4 of 7 agent communication protocols by recommenders
Open protocol that lets AI agents generate interface components as structured output, praised for rapid prototyping. Critics doubt it works beyond simple cases and warn of security and usability risks.
Open protocol that lets AI agents generate interface components as structured output, praised for rapid prototyping. Critics doubt it works beyond simple cases and warn of security and usability risks.
18 people gave an opinion
- recommend it
- 10
- criticize it
- 8
- approval share
- 56%
Recommended in 4 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 10 people recommend itNeeds 10 or more, met.
- 1.2 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 4 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Novel live AI-controlled interfaces
“The central breakthrough of this implementation is the move from static, database-driven visualizations to living, AI-controlled interfaces”
Comment from 15 Jan 2026 on Hacker News “I can justify A2UI as doing something not otherwise accomplishable in the market today”
Comment from 18 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” Pragmatic structured output for LLMs
“A2UI is most interesting for it's LLM generation options (not tools but structured output)”
Comment from 18 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” “a good and pragmatic way to approach the use of LLM systems”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” Clean, safe-looking design
“I quite like the look of this one”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” “Clean and safe but constrained.”
Comment from 20 Mar 2026 in “I turned Markdown into a protocol for generative UI”
What people warn about
Each mark is one comment
Needs language-specific libraries and components
“It still needs language-specific libraries”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” “they require building net-new components”
Comment from 28 Mar 2026 in “Show HN: ACP – An open protocol for agents to operate live UIs natively” Duplicates what HTML already does
“it's just a repackaging of the same problem, except in this case, the problem is solved with APIs and CLI”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” “it will eventually converge to what we already have, html”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” Limited beyond simple use cases
“Maybe this works for simple use cases but beyond that I'm skeptical.”
Comment from 16 Dec 2025 in “A2UI: A Protocol for Agent-Driven Interfaces” “Clean and safe but constrained.”
Comment from 20 Mar 2026 in “I turned Markdown into a protocol for generative UI”
Other agent communication protocols people recommend
The four most recommended other agent communication protocols, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Murmur by InstaVM | 14 | 0 | 100% | 6 | — | Approved |
| A2A (Agent2Agent) protocol | 14 | 21 | 40% | 5 | — | — |
| GitAgent | 11 | 4 | 73% | 1 | — | — |
| A2UIThis page | 10 | 8 | 56% | 4 | — | — |
| MCP Apps | 10 | 3 | 77% | 5 | — | Approved |
What people said
All 19 comments that mention it, newest first. Each links to the original comment.
“they require building net-new components”
“Clean and safe but constrained.”
“the models have to easily understand them or else you have to take up additional context explaining the protocol”
“The central breakthrough of this implementation is the move from static, database-driven visualizations to living, AI-controlled interfaces”
“I can justify A2UI as doing something not otherwise accomplishable in the market today”
“A2UI is most interesting for it's LLM generation options (not tools but structured output)”
“I personally think that's a non-starter”
“it's just a repackaging of the same problem, except in this case, the problem is solved with APIs and CLI”
“a good and pragmatic way to approach the use of LLM systems”
“But long-term it's probably training wheels.”
“it will eventually converge to what we already have, html”
“it only presents an actual UI 1/3rd of the time”
“This feels like a babydoggle to me.”
“It still needs language-specific libraries”
“for rapid prototyping this seems great”
“You're just asking for security bugs, UI impersonation attacks, terrible usability”
“I quite like the look of this one”
“Maybe this works for simple use cases but beyond that I'm skeptical.”
“This is very interesting if used judiciously”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.