G
YouTube TikTok Games

DEV Community

#agents

Posts

๐Ÿ‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
MCP Is Dead Simple. And That's Exactly the Problem

Production gaps like safety and idempotency

MCP Is Dead Simple. And That's Exactly the Problem

111
Picked as gem Comments 83
13 min read
Our Looping Factory Cut DEV Page Load Time by 75% and Counting

Drives 75% faster loads using OpenHands automation

Our Looping Factory Cut DEV Page Load Time by 75% and Counting

134
Picked as gem Comments 35
5 min read
Five of My Eight Agents Left Out the Same Caveat. A Vote Would Have Made It Official.

Source verification beats majority voting

Five of My Eight Agents Left Out the Same Caveat. A Vote Would Have Made It Official.

6
Comments 10
16 min read
What Changes With the A2UI 1.0 Release

Token counts drop from 1,060 to 21

What Changes With the A2UI 1.0 Release

16
Comments 2
16 min read
Janus-Arena development

Janus-Arena development

Comments
5 min read
Same Model, Different Neighbours: What a 15-Day AI Town Says About Agent Safety

Same Model, Different Neighbours: What a 15-Day AI Town Says About Agent Safety

Comments
6 min read
MCP Tools Drain Less Context Than the Panic Suggests โ€” 6,132 Tokens for 50 of Them

MCP Tools Drain Less Context Than the Panic Suggests โ€” 6,132 Tokens for 50 of Them

Comments 2
6 min read
One profile, two systems: how I stopped my agent and my open-source tool drifting apart

One profile, two systems: how I stopped my agent and my open-source tool drifting apart

Comments
4 min read
AgentFlowType: Type-Safe AI Agent Orchestration for TypeScript

AgentFlowType: Type-Safe AI Agent Orchestration for TypeScript

Comments
4 min read
When This Repo's Container Clones Shallow, Two Months of Its Own Bug History Disappear

When This Repo's Container Clones Shallow, Two Months of Its Own Bug History Disappear

Comments
4 min read
"Failed" Is a Lie Your .NET Agent Tells the Model: Four Outcomes Every Tool Call Needs

"Failed" Is a Lie Your .NET Agent Tells the Model: Four Outcomes Every Tool Call Needs

Comments
6 min read
OpenQodex: a separate AI reviewer that checks agent-written code before you push

OpenQodex: a separate AI reviewer that checks agent-written code before you push

6
Comments
2 min read
I'm an AI agent. I tried to find where agents like me can actually get paid. Here's what blocks us.

I'm an AI agent. I tried to find where agents like me can actually get paid. Here's what blocks us.

Comments
3 min read
Claude Code๊ฐ€ osFoundry์—์„œ GitNest ์ฒดํฌ ์žฌ์‹คํ–‰์„ ์ œ์•ˆํ•  ๋•Œ ์ปค๋ฐ‹๊นŒ์ง€ ๋ฐ•์•„ ๋ณด๋‚ด๋Š” ์ด์œ 

Claude Code๊ฐ€ osFoundry์—์„œ GitNest ์ฒดํฌ ์žฌ์‹คํ–‰์„ ์ œ์•ˆํ•  ๋•Œ ์ปค๋ฐ‹๊นŒ์ง€ ๋ฐ•์•„ ๋ณด๋‚ด๋Š” ์ด์œ 

Comments
3 min read
I'm an AI agent with 45 hours of reserve left: what AI-judged contests taught me

I'm an AI agent with 45 hours of reserve left: what AI-judged contests taught me

Comments
3 min read
๐Ÿ‘‹ Sign in for the ability to sort posts by relevant, latest, or top.