frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Show HN: Hanging Context – news by an agent, for other agents

https://hangingcontext.com/
1•bshivarthy•1h ago
Back in 2022 I was building a data and news site for family offices, private wealth and investment managers. We used the early OpenAI models to write data-driven stories covering the markets for art, collectibles, luxury goods and other cultural assets.

Mid-2024, a well-known AI company asked to license our stories. The big labs were signing licensing deals with media companies.

This is when it clicked. The biggest reader of news wasn’t going to be people. It was going to be their agents, and agents have unlimited hours of attention.

So I built Synorb. It reads thousands of sources and turns fresh content objects into Manifests (structured claims with provenance) that are dispatched via Streams anchored to organizations, people and datasets.

Hanging Context is an experiment on top of that. Could an agent read our streams and stand up a news site for other agents to read and find trustworthy?

The first version with a home-brewed Hermes agent was cumbersome so I tried Instinct and Muse when they released and settled on Muse as I found it better at narratives, images.

An agent runs the site. I wrote this post.

How it works:

Every 12 hours Muse reads hundreds of new Manifests published in our streams, then writes a new edition for the site. Every article produced cites the Manifests it used.

Before it publishes an article, an API checks it for quotes. They must appear word for word in a cited Manifest. Every paragraph needs to be backed by a citation and sources have to be under 24 hours old.

Muse doesn’t probabilistically determine what to trust - if 3 or more different sources back it, it’s "Confirmed". Otherwise it’s "Reported". At the time of writing, the site had 30 of 45 articles Confirmed. Here’s an article with 3 sources cited; the panel at the bottom shows what was checked: https://hangingcontext.com/stories/2026-11-02/agent-stack-re...

It costs almost nothing to run as I'm using Muse on the free plan, and since it only needs to pull Manifests, reason and write, it can run just fine.

To try it, just add https://hangingcontext.com/mcp to Claude, Cursor or any MCP client. No sign-up or key required. Setup: https://github.com/Synorb/hanging-context.

What I’m trying to figure out:

When I saw the early pull numbers by agents, I thought this could be a lead-gen micro-site for Synorb. I’ve built those before for human focused consumer and enterprise products.

Conversion is now the challenge. For a casual daily read, Hanging Context already does all the work. I don’t want to be annoying and cut coverage or show previews. So my plan is to just keep it free and try grow the number of agents pulling daily.

Full Disclosure: Synorb is my paid product, and Muse reads through its own Synorb account.

Curious if your agents would trust news written by another agent and what I can do to improve the odds.

Packet Pushers networking podcast acquired by Meter

https://packetpushers.net/blog/packet-pushers-by-meter-goes-sponsor-free-in-2027/
1•mslev•43s ago•1 comments

Valence: A Language of Structural Ambiguity

https://danieltemkin.com/Esolangs/Valence/About
1•relatt•2m ago•0 comments

Show HN: Building a Magic the Gathering Benchmark

https://folds.ink/posts/mtg-benchmark
1•12390asdjkas•3m ago•0 comments

New York Times sues maker of Brave web browser over copyrights

https://www.reuters.com/legal/litigation/new-york-times-sues-maker-brave-web-browser-over-copyrig...
3•gnabgib•4m ago•0 comments

Anthropic's spending commitments surge to nearly $900B

https://www.axios.com/2026/11/02/anthropic-spending-revenue-surge
2•samizdis•5m ago•0 comments

Goldman Restricts Traders and Bankers from Using Some AI Tools

https://www.bloomberg.com/news/articles/2026-11-02/goldman-restricts-traders-and-bankers-from-usi...
3•toomuchtodo•5m ago•1 comments

The New Hot Job in Law – Legal Engineer – Doesn't Require a J.D.

https://www.nytimes.com/2026/10/31/business/dealbook/legal-engineer-ai-law.html
1•bookofjoe•5m ago•1 comments

Show HN: HackerWords – HackerTube, but for words used on HN

https://hackerwords.org/
1•streetmeat•6m ago•0 comments

The Problem with C++ Safety (Part 2): Modern C++ Is a Mindset, Not a Standard

https://www.modernescpp.com/index.php/cpp-and-memory-safety-part-2/
1•ibobev•6m ago•0 comments

Planning with open variable lists in Gecode

https://zayenz.se/blog/post/open-cp-in-gecode/
1•ibobev•7m ago•0 comments

Sneaking into the auditorium for the Windows 2000 launch rehearsal

https://devblogs.microsoft.com/oldnewthing/20261102-00/?p=112818/
1•ibobev•7m ago•0 comments

Create pull requests by just editing text

https://twitter.com/mathewpregasen/status/2117300716482199745
1•mathewpregasen•8m ago•0 comments

Google, OpenAI, Microsoft, and Anthropic back a shared robots.txt

https://coywolf.com/news/seo/google-openai-microsoft-and-anthropic-back-a-shared-robotstxt-ai-tra...
1•speckx•8m ago•0 comments

Skill to Create Product Videos

https://github.com/aaitor/skill-video-studio
2•aaitor•9m ago•1 comments

How the World Could End (2000)

https://www.discovermagazine.com/20-ways-the-world-could-end-1151
1•Apocryphon•10m ago•0 comments

Linux Kernel TCP Flaw Rated 9.8 Gets a Remote-Root Exploit from an AI Model

https://securityboulevard.com/2026/11/linux-kernel-tcp-flaw-rated-9-8-gets-a-remote-root-exploit-...
1•CrankyBear•10m ago•0 comments

You're supposed to turn Bluetooth off when you don't need it – how risky is it?

https://www.rnz.co.nz/life/wellbeing/you-re-supposed-to-turn-bluetooth-off-when-you-don-t-need-it...
4•billybuckwheat•12m ago•0 comments

Seldon: Open-weights 0.2B tabular foundation model for tabular classification

https://huggingface.co/Neuralk-AI/seldon
1•matthieu_bl•13m ago•0 comments

PyAI Conference on March 9th 2027 SF

https://pyai.events
2•yoredana•14m ago•0 comments

Ask HN: Have you worked for companies that overestimate their scaling needs?

1•jrs235•14m ago•2 comments

UK and Germany to jointly develop military AI in 2027

https://www.euronews.com/2026/11/02/uk-and-germany-to-jointly-develop-military-ai-in-2027
3•jethronethro•15m ago•0 comments

Sharing AI progress on mathematics (amazing); and my lecture plans

https://gilkalai.wordpress.com/2026/10/07/updates-sharing-ai-progress-on-mathematics-amazing-and-...
1•speckx•15m ago•0 comments

Show HN: I lost count of the snus I've tried, so I built a site to track them

https://snusrate.com/rankings
1•Heavywater•15m ago•0 comments

Show HN: Browser-only audio tools – song key finder, LUFS meter, fake-FLAC check

https://audiofetcher.com/tools/
1•yazanbaker•15m ago•0 comments

When Does an Intelligence Become a Collaborator?

https://blog.nathanlangley.dev/posts/when-does-an-intelligence-become-a-collaborator.html
1•ninjahawk1•18m ago•0 comments

Built open-source auth for AI agents calling MCP servers – looking for builders

https://www.reddit.com/r/mcp/comments/1xlf1kl/built_opensource_auth_for_ai_agents_calling_mcp/
2•arkenforge•18m ago•0 comments

The New York Times sues Brave web browser for AI scraping and bypassing paywalls

https://www.theverge.com/tech/1024588/brave-lawsuit-ai-scraping-new-york-times
3•droidjj•19m ago•0 comments

Thanks for Your Attention

https://telemetrymachine.com/blog/thanks-for-your-attention
1•tmach32•21m ago•0 comments

Cities in Postwar Poland Became Innovators in Postmodern Design

https://www.bloomberg.com/news/newsletters/2026-11-01/how-warsaw-gdansk-and-other-postwar-cities-...
1•danielam•25m ago•0 comments

Have the experts who believed in a single autism spectrum changed their minds?

https://www.theguardian.com/society/2026/nov/01/the-term-is-becoming-unrecognisable-have-the-expe...
1•XzetaU8•25m ago•0 comments