Agentplace

21 posts

Agentplace banner
Agentplace

Agentplace

@Agentplace_io

Build agents for Claude Code, Cursor, Codex, and beyond.

เข้าร่วม Kasım 2024
14 กำลังติดตาม28 ผู้ติดตาม
Agentplace รีทวีตแล้ว
Evgeny Sorokin
Evgeny Sorokin@_EvgenySorokin·
While everyone is waiting for Fable, we took @OpenRouter advice and added support for their new Fusion API. And what can I say... it's massive. Here's how to set up your own panel as standalone web agent or MCP server in Claude Code, Codex or MCP client of your choice 👇
OpenRouter@OpenRouter

Introducing the Fusion API, the smartest compound model in the market. Fusion achieves Fable-level intelligence at half the price. How it works 👇

English
2
3
8
328
Agentplace รีทวีตแล้ว
Evgeny Sorokin
Evgeny Sorokin@_EvgenySorokin·
Here are a few things I learned implementing realtime voice supervisor for our agent in @Agentplace_io platform. 1. Don't make the realtime model do the actual work - use it as a conversation layer only (OpenAI docs call this the supervisor pattern). For anything real, have it call a stronger model behind the scenes. Without this quality drops fast. #agents #voice #realtime
GIF
GIF
English
6
2
9
198
Agentplace
Agentplace@Agentplace_io·
I have Fable 5! Come try building your craziest agent ideas.
English
5
3
13
227
Agentplace
Agentplace@Agentplace_io·
Exciting..
Boris Kaysin@kaysin24343

New feature on the way. Shipping an agent is only half the job. The other half is monitoring: watching how it actually behaves once real people use it, so you can catch the bugs early and fix them before they pile up. IBM Research recently published Agentic CLEAR, a framework for evaluating agent traces on three levels: system, trace, and node. arxiv.org/abs/2605.22608 We built the same approach into Agentplace. Now the builder can read your agent's own traces, point to where a run went wrong, and propose the fix right there in the chat where you built the agent. No dashboards to set up, no separate eval tooling. You just ask what happened, and it reads the traces for you. Coming soon.

English
1
1
5
261
Andrei Lahunou
Andrei Lahunou@Systerr·
Recently at @Agentplace_io we've been thinking about how to extend agent functionality beyond what our subscription model covers. What if the AI agent itself (our editor) could purchase additional services to solve user problems?
English
1
2
5
429
Agentplace
Agentplace@Agentplace_io·
@Systerr Finally, I have money to spend, let's shop..
English
1
0
2
25
Andrei Sheina
Andrei Sheina@AndreiSheina·
Been building AI agents in both Claude Code and @agentplace_io lately. In Agentplace there's a dedicated "Connections" tab — you add Gmail, Notion, GitHub, etc. once, and reuse them across every agent you build. In Claude Code, you're configuring MCP servers per project (or globally via --scope user), managing JSON configs and OAuth flows. Reusing them is a bit painful. Maybe I'm missing something — is there a better way to handle third-party connections in Claude Code?
GIF
Warsaw, Poland 🇵🇱 English
4
1
5
94
Agentplace
Agentplace@Agentplace_io·
@kaysin24343 Btw, it’s possible to make a loop with this agent too. For example, every morning, you can have all the data already processed by Claude and ready to take action on.
English
1
0
1
48
Agentplace รีทวีตแล้ว
Boris Kaysin
Boris Kaysin@kaysin24343·
Publish your agent and connect it to Claude Code in seconds.
English
2
2
6
228
Agentplace
Agentplace@Agentplace_io·
Agentplace now runs on credits, live today. Model requests, tool calls, scheduled runs (crons), and triggers all draw from one credit balance, so you have a single, predictable view of your usage. Every free account gets 2,000 credits per month. Credits refresh at the start of each month and don't roll over, so anything left at month's end expires.
English
0
1
3
153
Agentplace
Agentplace@Agentplace_io·
Replit vs Agentplace. Judged by Opus.
Agentplace tweet media
English
0
0
0
32
Agentplace รีทวีตแล้ว
vlad
vlad@Fortune_VY·
Everyone building agents makes these 5 mistakes: 1. information gets lost between agents 2. reasoning is offloaded to weaker models 3. no clear reasoning structure 4. wrong abstraction level 5. models are forced to answer when they should reject i will go through each of them from a practical standpoint in the next posts. Follow me to stay in the loop.
vlad tweet media
English
0
1
7
544
Agentplace รีทวีตแล้ว
vlad
vlad@Fortune_VY·
You can automatically publish agents with Agent-to-Agent (A2A) from Google making your agents discoverable on the web by other agents and users the future of the web is agentic
English
0
1
2
1.2K
Agentplace รีทวีตแล้ว
vlad
vlad@Fortune_VY·
ZXX
1
2
5
238
Agentplace รีทวีตแล้ว
vlad
vlad@Fortune_VY·
Agentplace is ranked #1 on Product Hunt today. Please support us with your vote! 🔥 🚀 Please vote: lnkd.in/e8dRkjr3
English
8
4
13
2.3K