Hacker News: harvenstar

Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents

harvenstar — Thu, 19 Mar 2026 16:36:09 +0000

Article URL: https://github.com/agentlayer-io/AgentClick

Comments URL: https://news.ycombinator.com/item?id=47442118

Points: 4

# Comments: 0

Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents

harvenstar — Mon, 16 Mar 2026 19:38:01 +0000

Article URL: https://github.com/agentlayer-io/AgentClick

Comments URL: https://news.ycombinator.com/item?id=47403743

Points: 1

# Comments: 0

New comment by harvenstar in "Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents"

harvenstar — Fri, 13 Mar 2026 18:38:05 +0000

I built AgentClick because we kept getting burned by AI agents acting too fast. The pattern is always the same: you ask Claude Code or Codex to do something, it generates a response, you get a y/n prompt in the terminal, and you approve it without really reading it. Then it sends the wrong email, runs a destructive command, or commits to a plan you'd never agree to if you actually saw it laid out.

The core problem: terminal y/n is not a real review step.

AgentClick adds a browser-based review layer between "agent proposes" and "agent executes." The agent drafts something, a UI opens in your browser, you can actually read, edit, and approve it visually — then the agent continues.

It works as a skill/plugin for: - Claude Code - Codex - OpenClaw - Any agent that can call HTTP tools

What you can review: - Email drafts and inbox triage - Shell commands before execution - Multi-step plans - Memory updates

The key difference from just adding a confirmation step: you can actually edit the content, not just approve/reject. Change the tone of an email, fix a command flag, remove a step from a plan — then let the agent proceed.

Install: `npm install -g @harvenstar/agentclick`

Works locally on localhost, or use `--remote` for a Cloudflare tunnel to review from your phone.

MIT licensed, open source: https://github.com/agentlayer-io/AgentClick

Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents

harvenstar — Fri, 13 Mar 2026 18:37:27 +0000

Article URL: https://github.com/agentlayer-io/AgentClick

Comments URL: https://news.ycombinator.com/item?id=47367965

Points: 2

# Comments: 1

New comment by harvenstar in "Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents"

harvenstar — Wed, 11 Mar 2026 22:32:43 +0000

Thanks for support!

New comment by harvenstar in "Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents"

harvenstar — Wed, 11 Mar 2026 18:10:58 +0000

I built AgentClick because I kept getting burned by AI agents acting too fast.

The pattern is always the same: you ask Claude Code or Codex to do something, it generates a response, you get a y/n prompt in the terminal, and you approve it without really reading it. Then it sends the wrong email, runs a destructive command, or commits to a plan you'd never agree to if you actually saw it laid out.

The core problem: terminal y/n is not a real review step.

It works as a skill/plugin for: - Claude Code - Codex - OpenClaw - Any agent that can call HTTP tools

What you can review: - Email drafts and inbox triage - Shell commands before execution - Multi-step plans - Memory updates

Install: `npm install -g @harvenstar/agentclick`

Works locally on localhost, or use `--remote` for a Cloudflare tunnel to review from your phone.

MIT licensed, open source: https://github.com/agentlayer-io/AgentClick

Show HN: AgentClick – Human-in-the-loop review UI for AI coding agents

harvenstar — Wed, 11 Mar 2026 18:10:37 +0000

Article URL: https://github.com/agentlayer-io/AgentClick

Comments URL: https://news.ycombinator.com/item?id=47339093

Points: 3

# Comments: 4

New comment by harvenstar in "Show HN: I built a site where strangers leave kind voice notes for each other"

harvenstar — Tue, 10 Mar 2026 22:37:56 +0000

Glad it clicked! Good luck with the project.

New comment by harvenstar in "Tell HN: It's official, I'm done with Claude"

harvenstar — Tue, 10 Mar 2026 22:11:46 +0000

I've been building a review UI layer for coding agents (Claude Code, Codex) that lets you actually inspect and edit what the agent is about to do before it executes: https://github.com/agentlayer-io/AgentClick

Turns out most of the "dumb" mistakes OP is talking about are catchable — you just need to actually see them before they ship.

New comment by harvenstar in "Tell HN: It's official, I'm done with Claude"

harvenstar — Tue, 10 Mar 2026 22:05:56 +0000

This resonates. The problem I keep running into isn't that the model is bad — it's that the feedback loop is too thin. A y/n in the terminal isn't enough to catch when the model does something subtly wrong.

New comment by harvenstar in "Launch HN: RunAnywhere (YC W26) – Faster AI Inference on Apple Silicon"

harvenstar — Tue, 10 Mar 2026 21:47:10 +0000

Cool project — been looking for something like this. Just opened a PR with a couple of new macOS actions (empty_trash + toggle_do_not_disturb). Happy to contribute more and quick chat if you're open to it.

New comment by harvenstar in "Show HN: I built a site where strangers leave kind voice notes for each other"

harvenstar — Tue, 10 Mar 2026 21:04:00 +0000

Maybe a middle ground — let people pick a pseudonym and a one-line tagline, but no real identity. You keep the warmth of anonymity while still letting listeners find voices they connected with.