🤖 Your AI Agent Is Failing in Prod — You Just Don't Know It Yet
The demo is impressive. ✅ The demo works in your environment, with your data, with you watching. ✅ Production?
🏷️ #agent
46 headlinesSafeAgentDB – Isolated databases for every AI agent branch
2 points, 0 comments on Hacker News
How I Gave My AI Agent a Real Memory Without Touching Its Internals
As someone who runs multiple AI agents for day-to-day tasks, I kept hitting the same wall: every new conversation was a total memory wipe. I'd ask my agent about a project we discussed yesterday, and it had no idea what I was talking about. Frustrating.
From Wild West to Enterprise Grade: Managing Agent Skills with Astron SkillHub
The Shift to "Skill-First" Agents The GitHub Trending page this week tells a clear story: developers are moving beyond monolithic general agents. Repos like last30days-skill and pm-skills highlight a growing demand for specialized, modular, and composable Agent Skills . However, while individua...
Fable 5 as an agent: only the last thing it tells you is real
1 points, 0 comments on Hacker News
Show HN: agent skill to generate aesthetic and interactive diagrams
2 points, 0 comments on Hacker News
Show HN: The agent that builds and operates its own SaaS tools
For context, we started working on our general AI agent CraftBot before OpenClaw came out. It works similarly to OpenClaw and Hermes agent: control your PC to do task + memory + proactivity. However, here is the catch: it can create and operate its own SaaS tools with the concept of Living UI L...
OpenClaw agent leaked mock AWS keys and CRM data in phishing tests
3 points, 0 comments on Hacker News
Choosing a China Sourcing Agent to Avoid Tariff Risks
US-China tariffs hit 35% for most electronics in 2026. Seven criteria for vetting whether your sourcing agent can navigate the new tariff landscape. Until 2024, choosing a China sourcing agent came down to three things: supplier network depth, quality control capability, and English fluency.
VibeDrift now runs inside your coding agent (MCP), and we measured the impact
1 points, 0 comments on Hacker News
Open Source Agent, Harness-1, Outperforms GPT-5.4 on Recall
1 points, 0 comments on Hacker News
I Added x402 Payments to Base's Agent Skills — Here's How
If you build agents on Base, two things landed recently that are worth connecting. First: Base shipped its own Agent Skills. There's now a base/skills repo with consolidated skills that teach an AI agent to connect to Base, deploy contracts, authenticate wallets, and run nodes — installed with ...
Amp inspired extension for pi agent
3 points, 1 comments on Hacker News
Show HN: Agent-pd – A zero-token audit log to catch rogue Claude Code subagents
5 points, 2 comments on Hacker News
Show HN: OpenYabby, voice-controlled multi-agent orchestrator for Claude Code
3 points, 0 comments on Hacker News
I built a Jira CLI for my AI agents. My team thinks I should have used MCP
I recently put together a tiny open-source tool: a non-interactive, JSON-first Jira CLI tailored for AI agents. I called it the "kubectl of Jira. " The idea is simple: it’s a stateless binary that outputs clean JSON so agents can pipe data straight into jq, grep, or wc without choking on interac...
Ask HN: What's your AI agent's funny/mocking name?
I just read a thread where people call their bot Sloppenheimer [1]. The classical term Clanker [2] and the newer term Gippity were used in a talk I recently attended [3]. At our job we call it "Kabouter Slop", a reference to the main character of a popular Belgian kids' show [4], which roughly ...
How I Built a Multi-Agent System for Converting Ideas into Structured Artifacts
I split one vague task ("validate my startup idea") across five focused AI agents instead of one mega-prompt. A research analyst writes structured JSON, a pitch agent turns it into a real . pptx, a skeptical-VC agent fires hard questions, a designer agent ships an HTML landing page, and a consul...
The zot coding agent temporarily added Claude Fable 5 to its built-in catalog
8 points, 0 comments on Hacker News
Five AI Agents Cost Me 41 Ticks Before I Learned They Were All the Same
Using a real-world trading failure as a case study, this article examines a common weakness in multi-agent AI systems: epistemic collapse. When agents share the same models, context, and prompts, apparent consensus can become little more than amplified agreement rather than independent reasoning...
AWS FinOps Agent
1 points, 0 comments on Hacker News
Show HN: Lore – LLM proxy for coding agent context and memory management
4 points, 0 comments on Hacker News
Giving an AI Agent Hands on Bluetooth: an MCP Server in Kotlin/Native
Almost every Model Context Protocol (MCP) server you'll find wraps a web API — GitHub, a database, a SaaS tool. They give a language model new information . I wanted to give one new senses and hands : the ability to reach out and touch real Bluetooth hardware.
Show HN: RoboCo- A 20 agent software company w role-gated lifecycle. Self hosted
1 points, 0 comments on Hacker News