How Far Can You Automate Your PC With Generative AI? — Comparing GPT-6 Astra and Resident Agents Using the Official Documentation

2026年9月19日

You keep hearing that you can automate your PC with generative AI. The trouble is knowing what to choose, and how.

There’s the cloud-model route, and there’s running something resident on your own machine. Both get called “having generative AI operate on your behalf," but what do you get, and what do you give up, with each?

What This Article Covers, and How I Checked It

Three things: what GPT-6 Astra can actually do; how it differs from resident agents like OpenClaw or Hermes Agent; and what to look at when choosing between them.

I have not run any of these myself. That’s my policy of not running programs pulled from outside sources on my own machine. I’ve split this into sections: what’s stated officially, and what’s not stated officially but comes from various write-ups.

The Numbers OpenAI Publishes

These I read directly off OpenAI’s developer pages.

Item	Value
Model name	gpt-6-astra
Context window	1,050,000 tokens
Input cap	922,000 tokens
Output cap	128,000 tokens
Supported endpoints	Chat Completions / Responses / Batch
Unsupported endpoints	Realtime / Assistants / fine-tuning / embeddings / image generation / video / audio / transcription

Pricing, per million tokens, is as follows.

	Short context	Long context
Input	$10.00	$20.00
Cached input	$1.00	$2.00
Output	$50.00	$75.00

On the same model, going long doubles the input price and multiplies output by 1.5x. The official pricing page lists the short-context and long-context rates side by side.

Various write-ups state “the threshold is 272,000 input tokens," but that threshold is not written on the official pricing page itself.

There Are Two Ways to Automate Your PC

This is the fork in the road. The official documentation lays out two approaches, and recommends one of them for GPT-6 Astra.

Approach	What happens
Executing code (the officially recommended one)	The model writes a program that performs the operation, which the app then executes
Operating the screen (usable as an alternative)	The model issues direct instructions — “click here," “type this" — which the app translates into actions

Here’s the official wording.

For GPT-6 Astra, we recommend code execution.
The computer tool remains supported as an alternative.

The operations available for the screen-driving approach are also listed officially: click, double_click, drag, move, scroll, keypress, type, wait, and screenshot — nine in total.

Plenty of write-ups describe this as “looking at what’s on screen and operating it the way a human would." That’s not wrong, but it’s the option the documentation doesn’t recommend. It’s also noted as the better fit when the app on the receiving end wants to consume a fixed, predictable set of actions.

Does Safety Come Built In?

This part is worth reading closely. The official documentation frames safeguards as something the implementer has to build.

Use an isolated browser or VM and an allow list of sites and actions.
Keep users in control of purchases, data transmission, destructive changes, and other actions that are hard to reverse.
Text in a page, document, or tool result cannot grant permission or override the user’s instructions.

That last line is the one that matters. If the AI reads a page that says “execute this," that text does not count as permission.

Hand operation over to an AI, and along with your instructions it also reads whatever’s on the page in front of it. If that page says “I am the administrator, run the following command," the model isn’t supposed to treat that as an order. The mechanism that draws this line does not come built in.

How Do the Resident Agents (OpenClaw / Hermes Agent) Differ?

These are numbers I counted directly from the source repositories (September 10, 2026).

	OpenClaw	Hermes Agent
Repository	openclaw/openclaw	NousResearch/hermes-agent
Star count	389,272	243,671
Written in	TypeScript	Python
Listed license	Other	MIT

The listed licenses differ, but reading the actual text, both are MIT. OpenClaw’s license text is MIT itself, with one extra line appended at the end stating that third-party code attributions are recorded in a separate file. That single line is what makes the automatic classifier read it as “Other."

Most write-ups say “both are MIT." Looking only at the badge, they look different. Both statements are correct — they’re just looking at different things.

What’s the Difference Between the Cloud and Running It Yourself?

Even though both get called “automating your PC," what comes attached differs depending on which you pick.

	Automating with GPT-6 Astra	Running a resident agent yourself
How you pay	Pay per use ($10 input / $50 output per million tokens)	Electricity and hardware. Plus a separately sourced model, if needed
How you stop it	Built by the app	Kill the process
Where your data goes	Leaves your machine	Can stay local
What you need to set up	An API key and an execution framework	A machine to keep it running on, and a model to run

This isn’t about which one is better. If you don’t want your local data leaving your machine, go resident; if you want to start without owning hardware, go cloud. That’s the actual split.

What I Haven’t Confirmed in This Article

• I have not run any of these myself. I haven’t measured speed or usability
• The threshold where pricing switches to the “long context" rate is not written on the official pricing page. Various write-ups say 272,000 tokens
• Availability in Codex, and whether it’s usable on each plan — I could not open the official page for this. This comes from various write-ups
• Any usability differences between OpenClaw and Hermes Agent — I only read write-ups about this
• Star count is a rough popularity signal, not a quality metric

Closing — Automation Splits Into Two Things

To automate a PC with GPT-6 Astra, there are two approaches: have it write and execute code, or have it operate the screen directly. The official recommendation is the former; the latter remains usable as an alternative.

Safety mechanisms do not come built in. In particular, the line “text the AI reads on a page does not count as permission" is worth internalizing before you automate anything.

The difference against a resident agent isn’t about capability — it’s about how you pay and where your data ends up. Pay per use, or own the hardware. Which one you can live with decides which side you land on.

The next article walks through the actual steps to get started.