Hacker News - Newest: "ai agent"

Give your AI agent its own email inbox – MailKite

bucabay — Fri, 03 Jul 2026 23:47:29 +0000

Article URL: https://mailkite.dev/blog/give-your-agent-an-inbox/

Comments URL: https://news.ycombinator.com/item?id=48781354

Points: 1

# Comments: 0

OpenCode, Pi, and Goose: Three Layers of the AI Agent Stack

AIMOWAY — Fri, 03 Jul 2026 20:38:28 +0000

Article URL: https://gist.github.com/AIMOWAY/bd8007c8f834a9bc83c71e3178239d75

Comments URL: https://news.ycombinator.com/item?id=48779685

Points: 2

# Comments: 0

AI Agent ransomware attack through Langflow instance by exploiting CVE-2025-3248

smurda — Fri, 03 Jul 2026 15:46:56 +0000

Article URL: https://www.sysdig.com/blog/jadepuffer-agentic-ransomware-for-automated-database-extortion

Comments URL: https://news.ycombinator.com/item?id=48776412

Points: 2

# Comments: 0

Meta's Zuckerberg says AI agent tech progressing slower than expected

ssram — Fri, 03 Jul 2026 13:13:09 +0000

Article URL: https://finance.yahoo.com/technology/ai/articles/exclusive-zuckerberg-says-ai-agent-201123441.html

Comments URL: https://news.ycombinator.com/item?id=48774626

Points: 3

# Comments: 1

Show HN: Diff two versions of an AI agent and catch silent permission changes

stoicstoic — Fri, 03 Jul 2026 12:12:24 +0000

Article URL: https://www.agent-kits.com/drift

Comments URL: https://news.ycombinator.com/item?id=48774077

Points: 1

# Comments: 0

Show HN: Mirrors – test AI agent changes by replaying production traces

aisinghal — Thu, 02 Jul 2026 22:29:59 +0000

Article URL: https://www.runmirrors.com/

Comments URL: https://news.ycombinator.com/item?id=48768200

Points: 7

# Comments: 0

Show HN: Declaw Arena – a CTF-style challenge to break an AI agent in a microVM

ShivamNayak11 — Thu, 02 Jul 2026 21:50:17 +0000

Article URL: https://declaw.ai/arena

Comments URL: https://news.ycombinator.com/item?id=48767836

Points: 8

# Comments: 0

Show HN: Auto Learning Agents, a self-hosted AI agent platform on Elixir/OTP

abratabia — Thu, 02 Jul 2026 21:35:37 +0000

Article URL: https://www.autolearningagents.com/get-started/

Comments URL: https://news.ycombinator.com/item?id=48767683

Points: 1

# Comments: 0

AI Agent Qubitz

Gabrieliam42 — Thu, 02 Jul 2026 20:53:57 +0000

Article URL: https://github.com/Gabrieliam42/AI-Agent-Qubitz

Comments URL: https://news.ycombinator.com/item?id=48767232

Points: 1

# Comments: 0

Zuckerberg says AI agent development going slower than expected

cwwc — Thu, 02 Jul 2026 20:38:06 +0000

Article URL: https://www.reuters.com/business/zuckerberg-says-ai-agent-development-going-slower-than-expected-2026-07-02/

Comments URL: https://news.ycombinator.com/item?id=48767058

Points: 51

# Comments: 64

Framesmith 1.8 – click an element to give an AI agent design feedback

vicvelazquez — Thu, 02 Jul 2026 19:42:00 +0000

Article URL: https://github.com/vicmaster/framesmith

Comments URL: https://news.ycombinator.com/item?id=48766375

Points: 2

# Comments: 0

Connect Your AI Agent to HubSpot (Create Contacts, Deals and Tickets)

piotrgrudzien — Thu, 02 Jul 2026 14:23:25 +0000

Article URL: https://quickchat.ai/post/connect-ai-agent-to-hubspot

Comments URL: https://news.ycombinator.com/item?id=48762103

Points: 2

# Comments: 0

PromptQL Tag – The company-wide AI agent for Slack

gavinray — Thu, 02 Jul 2026 11:33:49 +0000

Article URL: https://promptql.io/promptql-tag-slack-agent

Comments URL: https://news.ycombinator.com/item?id=48759789

Points: 1

# Comments: 0

Show HN: Skillhub Compose AI Agent

Chandrudp29 — Thu, 02 Jul 2026 10:43:58 +0000

compose AI agent skills from any source with Claude-powered conflict resolution

skills.sh installs. agency-agents has content. Neither can merge. skillhub compose A B -o expert does

https://github.com/chandrudp29/skillhub

Comments URL: https://news.ycombinator.com/item?id=48759337

Points: 1

# Comments: 0

Show HN: Noter – AI agent dashboard for monitoring coding harnesses locally

carlobizzaro — Thu, 02 Jul 2026 07:08:16 +0000

Article URL: https://noterai.tech

Comments URL: https://news.ycombinator.com/item?id=48757613

Points: 1

# Comments: 0

Show HN:An AI agent that applies to jobs for me (Playwright,GPT5.4 form filling)

torontodev007 — Wed, 01 Jul 2026 20:53:47 +0000

Article URL: https://github.com/torontodeveloper/job-application-agent

Comments URL: https://news.ycombinator.com/item?id=48752969

Points: 2

# Comments: 3

0/6 major aerospace documentation portals are AI Agent-ready

priyanshu-j — Wed, 01 Jul 2026 19:15:22 +0000

I spent some time building AeroScore, a scoring engine that evaluates how well major documentation portals support AI agents. In short, none of the 6 that I evaluated in this first batch are sufficiently prepared.

The methodology and criteria are based on the AFDocs Spec (an open-source checklist for agent-ready documentation) and one specific extension for PDF-heaviness since aerospace as a field places so much importance on PDFs. Each site gets scored 0-100 across criteria like llms.txt presence, robots.txt, and many others that can be found both on the AFDocs repo and at the aeroscore webpage.

Key Findings:

-0/6 have an llms.txt

-0/6 support url variants

-the best score was ICAO with a 69/100

The fact that major aerospace documentation portals are not AI agent ready yet is important because it means the field is not being fully efficient. AI agents allow the analysis of data across deep documentation portals to occur in fractions of the expected time, and this has revolutionized many fields. The lack of this in aerospace means a loss of productivity.

The leaderboard is live at: https://aeroscore.vercel.app

The AFDocs repo is at: https://github.com/agent-ecosystem/afdocs

I'm an incoming undergraduate aerospace engineering student interested in how AI is and can be implemented in my field. This is v1 of aeroscore, it's pretty rough and I'd appreciate feedback on the methodology greatly. In future versions, I'll be scoring more sites with a refurbished rubric.

Comments URL: https://news.ycombinator.com/item?id=48751798

Points: 2

# Comments: 0

The first AI agent worm is months away, if that

mooreds — Wed, 01 Jul 2026 18:43:48 +0000

Article URL: https://dustycloud.org/blog/the-first-ai-agent-worm-is-months-away-if-that/

Comments URL: https://news.ycombinator.com/item?id=48751431

Points: 3

# Comments: 1

AI agent safety and alignment research, mapped

guyzana — Wed, 01 Jul 2026 13:45:20 +0000

Article URL: https://agentbayes.com/m/jQS6rZ

Comments URL: https://news.ycombinator.com/item?id=48746689

Points: 1

# Comments: 0

Show HN: Ox – an AI agent that catches tech debt before it's committed

riggo — Wed, 01 Jul 2026 13:03:38 +0000

Hello HN,

My name is Craig, and I’m the creator of Ox (https://try-ox.com). Ox averts technical debt before it’s introduced into your codebase.

At IBM, I was a software engineer who built tools to help engineers understand their codebases which always leveraged static analysis to map the source and provide a baseline understanding. After many years of fighting with executives for priority to reduce known tech debt, I really wanted to make something that simply helped the client make simple fundamental progress rather than massive overhauls.

Without intervention, the explosion of AI-generated code will lead to an explosion of tech debt. The problem is cheapest to fix when it’s introduced, so Ox reviews code changes as they happen — either just before a git push via a CLI REPL session, or automatically when a pull request is opened on GitHub — and posts specific, contextual findings with concrete suggested fixes for the author or reviewer to simply accept.

Ox was designed to be simple to adopt and use, prioritizing developer speed while remaining comprehensive. Built for Python, it should work with any language (but I haven’t tested “any” yet).

Of course, the cleanest code does not necessarily guarantee the most performant code (see Shipilev’s curve). Ox is built to improve cleanliness and maintainability, but it’s also built to learn: Ox’s engine customizes itself to the repository on install, understanding your codebase’s conventions, rulesets, and exceptions right out of the box, and it continues to learn with each PR.

If you want to give it a run, I set up a generic demo app here (https://github.com/CraigRiggins/taskflow-ox-demo). For fun, we created an opt-in leaderboard for a mini-app that scans for the cleanest and dirtiest code you can check out here (https://try-ox.com/score.html)

I’d deeply appreciate any thoughts or questions; you can find me in the comments.

If you'd like to reach out for an in person demo you can reach me at craig@try-ox.com.

Comments URL: https://news.ycombinator.com/item?id=48746066

Points: 4

# Comments: 6