AI should do the implementation. You should own the decisions.

Cesar — Sat, 13 Jun 2026 09:59:45 +0000

The default for AI-assisted development is one of two failure modes.

Either you're babysitting the agent line by line — approving each diff, re-explaining context it dropped three messages ago — or you've handed it the wheel and you're hoping the PR that lands at the end resembles what you asked for.

Son of Anton is neither. It's a delivery orchestrator built on a single claim: there are exactly three moments where a developer's judgment is irreplaceable. The orchestrator owns everything in between.

The three gates

Every project moves through three human decision points. Nothing important happens without you signing off.

Gate 01 — Approve the WHAT (/soa plan)
A grill-me session forces the AI to surface its assumptions, constraints, and scope decisions back to you before a single ticket exists. You say yes or you refine. It does not proceed until you have.

Gate 02 — Approve the HOW (/soa decompose)
The approved plan becomes a ticket stack — ordered, dependency-aware, sized for review. Architectural judgment stays with you. Ticket authorship goes to the agent.

Gate 03 — Approve DONE (/soa closeout)
An adversarial subagent reviews every ticket before its PR opens. When the phase is complete, you decide whether to accept. Closeout squash-merges the stack onto main. Nothing merges without you.

Between the gates, you are not needed

That's the whole point. Once you've approved the plan and the tickets, the orchestrator runs the loop:

I gave my AI coding agent a pet. It reacts to everything it does.

Cesar — Fri, 12 Jun 2026 05:39:05 +0000

I spend all day pair-programming with an AI agent, and it felt a little lonely
staring at a terminal. So I built Codogotchi — a tiny chibi pet that lives in
your macOS menubar and reacts to what your coding agent is actually doing.

It's not a spinner. The pet has 19 animation states wired to real agent
lifecycle events — implementing, testing, thinking, reading, running git,
errored, waiting on you, standby. When your agent hits a wall, the pet looks
frustrated. When a PR lands, it celebrates.

It also has a pulse: your pet's health decays if you stop coding (with weekend
grace + a vacation mode so it doesn't guilt-trip you on holiday). Neglect it
long enough and it ghosts out — revive it by getting back to work.

Works with Claude Code, Codex, Cursor, Copilot, Antigravity — anything that can
run a hook.

A few things I'm weirdly proud of:

Hatch your own pet with AI: describe one or drop a seed image and your agent generates a full animated spritesheet, one row at a time.
A community gallery — install someone else's pet with npx codogotchi add <id>.
An RPG layer in progress: real coding activity earns XP and levels (no pay-to-win, ever).

It's free. Default pet is Maew, a Thai-inspired chibi (I'm building this from
Thailand ☕).

Try it: https://codogotchi.app

Genuinely want feedback — what would make you actually keep this in your menubar instead of closing it after a day? Roast welcome.

DEV Community: Cesar

AI should do the implementation. You should own the decisions.

The three gates

Between the gates, you are not needed

I gave my AI coding agent a pet. It reacts to everything it does.