agentdev001

5 ms·

1.
by agentdev001 3d ago
IIRC, openai enabled the ability for subagents to use different models than the parent- maybe a month~ ago.
2.
by agentdev001 5d ago
And replying with some more thoughts after reading the comments here. To me, this feels similar to when models started passing the line of (imo) "good enough" to start building much more capable agents. The release of this (gpt-li
3.
by agentdev001 5d ago
Reposting my comment from https://news.ycombinator.com/item?id=49646963 Congrats on the launch here. I've been messing with this over the last few hours- super super cool. I was excitedly awaiting this hitting the API,
4.
by agentdev001 6d ago
Native integration with their SDK.
5.
by agentdev001 6d ago
Congrats on the launch here. I've been messing with this over the last few hours- super super cool. I was excitedly awaiting this hitting the API, because ofc there wasn't a super high fidelity option for drop-in voice interface i
6.
by agentdev001 7d ago
Your comment frustrates me! With that said, how about Discord? I have high speed internet 90% of my waking life, high enough for Discord to be high fidelity. I actively avoid using discord to join calls on my phone- because Bluetooth absolu
7.
by agentdev001 8d ago
Until either of these cases happen: A: Devs with an online presence stop using Anthropic models B: Anthropic catches up to OpenAI in terms of per-token efficiency, and average token total for final-output We will continue to see posts such
8.
by agentdev001 11d ago
Secrets shouldn't be plain text in project directory >:[
9.
by agentdev001 13d ago
I am not using MCP in be production- but, my team is. My team also produces MCP servers for other teams, and I find it a bit maddening. I wonder if anyone can relate to my experience here. It feels like there is a significant amount of bagg
10.
by agentdev001 13d ago
> "it makes things a tiny bit easier than interfacing directly via api" By what metric? I would expect that a thin API client (with readable code) is generally going to out-perform a tool-surface which you don't have the a
11.
by agentdev001 13d ago
Sounds like you might benefit from running a custom harness then, no? I can't imagine for a task such as that- that codex is the best option.
12.
by agentdev001 15d ago
3.8 Flash tomorrow apparently, WSJ says google insiders say on-par with Opus 5. Time will tell.
13.
by agentdev001 15d ago
This is moreso about the (human-intended) tools, data, and environments you have available to you. Wanna do defense? Get more telemetry. Wanna do red? Get solid test-bed environments. Mature infosec programs are benefiting the most, good-gu
14.
by agentdev001 15d ago
High, X-high, and Max are all on the $/intelligence Pareto
15.
by agentdev001 15d ago
Could be risky. Yet goal solution.
16.
by agentdev001 16d ago
I think its moreso ChatGPT desktop is the platform, and the split is Chat, vs Codex/Work. Codex and Work being largely the same thing, but with UI and tool surface differences geared towards SWE vs Business work. But yes; terrible UX a
17.
by agentdev001 16d ago
Have you got some fancy scroll thing for your blog? Something in my brain is noticing that scrolling feels smoother than your average site, though maybe im crazy.
18.
by agentdev001 17d ago
Wow. For what it's worth, I ran that prompt in Codex mode in ChatGPT Desktop (sans-site artifact), and I ended up with the below. This is on a default setup, win10. - *9 top-level orchestration tools*: the `functions. ` and `collaborat
19.
by agentdev001 17d ago
Codex in ChatGPT Desktop + 5.6Sol is my daily driver for non-coding things, and it's great. FWIW, I've not explored what differentiates Codex and Work modes- Simon notes that Work 'feels more like regular Codex re-skinned&#x2
20.
by agentdev001 19d ago
Did you even review this? Holy moly. "I'm just trying to generate content that’s informative and useful, to someone, anyone". Come on now, you're joking right? "or send a paranoid infoSEC analyst into a full lockdo
21.
by agentdev001 21d ago
Well, effectively, yea. > remove alignment > give impossible task > actor exhausts all options possible within knowledge + toolset
22.
by agentdev001 23d ago
The post ends with a comment on "its not about being pedantic..." so, a few not being pedantic bits: In the table "Real world examples"; "Claude Desktop" houses three harnesses at the moment; Claude, Claude Cow
23.
by agentdev001 25d ago
Devil's advocate will say "Well, the agent would need an MCP client to use MCP-served resources... if you can give it that, why not give it an HTTP client?"
24.
by agentdev001 26d ago
Im trying to do my part in the world here. This post needs an edit. Author is not comparing "Codex" and "Claude". They are comparing Codex TUI/CLI with (presumably) gpt-5.6-sol, against Claude Code TUI/CLI with
25.
What Is a Harness?(earendil.com)
2 points by agentdev001 27d ago|0 comments
26.
by agentdev001 28d ago
Does a non-provider harness not offer this?
27.
by agentdev001 28d ago
OpenAI allows you to use the subsidized limits outside of their products. I don't think they clearly "tell" the world that this is the case, but they do put out a product surface that- I dont think they would, if they didnt a
28.
by agentdev001 28d ago
Shoot, yes. The user experience of logging in is the same, but yes- using a different (non-cc) harness bumps you up to the pay-per-token rates. However! I have seen projects which use CC under the hood, in order to get the subsidized rates.
29.
by agentdev001 28d ago
They did walk that back, you are able to use an Anthropic login in a non-anthropic harness. (For now.) (Edit: I failed too mention, using a different harness with a CC/Anthropic subscription bumps you up to the pay-per-token rate.)
30.
by agentdev001 28d ago
> less rambling and tangential This is not necessarily a problem with the harness (IE, swapping from Claude Code wouldn't necessarily fix this). If what you're looking to avoid is specifically 'rambling and tangential&#x27

More ›