frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Don't Wire Workflows, Build Skills Instead

https://github.com/MooseGoose0701/skill-compose
1•ChocoluvH•1h ago

Comments

ChocoluvH•1h ago
Don't Wire Workflows, Build Skills Instead!

Hi HN! I'm really excited to share this. I'm the creator of Skill Compose.

I've been an NLP engineer for almost 10 years, and I've been a workflow fan ever since Google Dialogflow.

THE WORKFLOW RENAISSANCE (AND WHY IT MADE SENSE IN 2023)

Tools like Dify, and later n8n pushed "agent building with workflows". In 2023, that made sense for two reasons:

1) LLMs were unstable at basic tasks, so you needed deterministic nodes for checks and branching.

2) Instruction following was weak (especially open source), so workflows enforced control flow.

Workflows exploded. For a while, they were the right tool.

BUT SOMETHING DIDN'T FEEL RIGHT

Over the last two years, LLMs have improved at a pace that still shocks me.

But workflows haven't.

It feels like we're building agents in 2026 with a technique from 2016.

As logic grows, workflows get expensive to build, hard to read, painful to maintain, and awful to migrate.

We called it "Artificial Intelligence", but really?

Artificial, yes. Intelligence, not so much.

WHY I THINK SKILLS WIN (EVEN IF YOU STILL LIKE WORKFLOWS)

1) Skills enforce best practices.

LLMs are smart, but they can't reliably infer how your company works, like your slide standards or your security boundaries.

A skill is where you encode that reality as a versioned, reviewable, executable spec.

2) Skills evolve by local edits, not orchestration rewires.

Most iteration is about raising the standard, not reshaping the process. With skills, you improve rubrics, examples, specs, and deterministic helpers.

3) Skills don't collapse when reality drifts off the "happy path"

Workflows handle the happy path. When inputs drift or something breaks, they need a growing tree of edge cases.

Skills are different. Modern LLMs can write, run, and revise code to diagnose failures and recover, so the system can self heal instead of dead ending.

SO, WHAT SKILL COMPOSE IS

Skill Compose is an open source platform for building and shipping skill powered agents, no workflow graphs, and no CLI.

THE REAL GOAL: SKILLS FOR EVERYONE

Skill-powered agents should be accessible to anyone who has a repeatable process worth encoding.

---

Design Principle #1: Container first runtimes + Executors

Skill Compose is container first by default: agents run in containers (or K8s pods), not on your host.

You can run locally, but once an LLM executes code you start worrying about deps, random binaries, and breaking your system. So each agent gets its own isolated runtime for safer, reproducible execution.

By default, agents use the standard Skill Compose image. For heavier needs (GPU, large ML stacks, HPC), you can assign an Executor: a custom Docker image or K8s pod, even if it's tens of GB with CUDA, frameworks, and domain dependencies.

---

Design Principle #2: Evolution via feedback + traces.

Often times we wanted to give skills explicit feedback + real run traces to propose upgrades prompts, examples, scripts.

Start mediocre. Improve fast.

---

Design Principle #3: Skill management.

How about importing skills directly from a GitHub repo URL?

Also support portable formats: .skill files, .zip archives, and drag and drop folders.

Then treat skills like real software: version them, review them, diff them, and fork them.

---

Design Principle #4: No "skill marketplace" (AND WHY)

Black box skills you do not own are rarely useful.

The skills that matter are the ones you build, adapt, and fully control. They encode your workflows, policies, data shapes, edge cases, and permission boundaries.

Ownership is what makes skills auditable, secure, and maintainable.

---

THE HONEST DISCLAIMER

This is still early. Some parts are rough. Some parts are toy like. A lot will change.

---

Thanks! And please be blunt. Critical feedback is the fastest way this gets better.

Show HN: FlareBar – Access your Cloudflare dashbord from macOS menu

https://flarebar.app/
1•mrbutttons•2m ago•0 comments

Rewriting an Objective-C project in Swift with the Xcode agent support

https://mastodon.social/@stroughtonsmith/116018205506714527
1•Austin_Conlon•3m ago•0 comments

Apple Confirms Revamped Siri Is Still Coming in 2026

https://www.macrumors.com/2026/02/12/siri-ios-26-launch-confirmed-apple/
1•newman314•4m ago•0 comments

ARC-AGI-1 and 2 LEADERBOARD

https://arcprize.org/leaderboard
1•doener•8m ago•1 comments

Show HN: Paragliding RL

https://southriverai.github.io/southriverblog/post.html?slug=the-speed-to-fly-in-2026
1•kozzion•9m ago•0 comments

Trump Justice Department Poised to Preserve Ticketmaster Monopoly

https://prospect.org/2026/02/12/trump-justice-department-ticketmaster-live-nation-monopoly/
1•leotravis10•10m ago•0 comments

Justice Department antitrust chief Gail Slater resigns

https://www.axios.com/2026/02/12/justice-antitrust-chief-gail-slater-resigns
1•leotravis10•11m ago•0 comments

Google may be cracking down on self-promotional 'best of' listicles

https://searchengineland.com/google-cracking-down-self-promotional-best-of-listicles-468227
1•gnabgib•14m ago•0 comments

Show HN: Sovereign Suite – A Recursive Logic Framework for AI Governance

https://github.com/holland202/Sovereign-Suite-Manifest
1•badatchess•18m ago•0 comments

Show HN: New Open Source Agent with 62 Stars on GitHub

https://github.com/dakotalock/holygrailopensource
2•Moriarty2027•22m ago•0 comments

Mitchell Hashimoto Launches 'Vouch' to Fight AI Slop in Open Source Ecosystem

https://itsfoss.com/news/mitchell-hashimoto-vouch/
2•WaitWaitWha•22m ago•1 comments

Ethnic minorities are driving America's startup boom

https://www.economist.com/finance-and-economics/2026/02/12/ethnic-minorities-are-driving-americas...
1•andsoitis•23m ago•0 comments

Authoring, simulating, and testing dynamic human-AI group conversations

https://research.google/blog/beyond-one-on-one-authoring-simulating-and-testing-dynamic-human-ai-...
1•gmays•24m ago•0 comments

PostgreSQL v19: Password expiration warnings

https://hexacluster.ai/blog/postgresql-v19-password-expiration-warnings
1•avivallssa•28m ago•0 comments

Show HN: Khaos – Every AI agent I tested broke in under 30 seconds

1•exordex•29m ago•0 comments

How Are Amps Modeled? [video]

https://www.youtube.com/watch?v=9YL8pwF7Mnc
2•dsego•32m ago•0 comments

What 1.4M emails reveal about America's most notorious sex offender

https://www.economist.com/interactive/international/2026/02/12/inside-epsteins-network
2•doener•33m ago•0 comments

Simile: The Simulation Company

https://twitter.com/joon_s_pk/status/2022023097017421874
1•jaehong747•35m ago•0 comments

Elide is an all-in-one, AI-native, open source software runtime

https://elide.dev/
2•shirian•37m ago•0 comments

The March Cliff: Why the 2026 Economic Collapse Is Different

https://ramakanth-d.medium.com/the-march-cliff-why-the-2026-economic-collapse-is-different-e1c619...
1•playhard•38m ago•2 comments

Welcome to the Great Regression

https://www.bloomberg.com/opinion/newsletters/2026-02-12/the-us-risks-a-great-regression
1•petethomas•39m ago•0 comments

Judge rules that LLM provided legal advice is open to discovery [pdf]

https://storage.courtlistener.com/recap/gov.uscourts.nysd.652138/gov.uscourts.nysd.652138.22.0.pdf
3•stingrae•40m ago•0 comments

My hot take on vibe coding for PMs

https://www.ddmckinnon.com/2026/02/11/my-%f0%9f%8c%b6-take-on-vibe-coding-for-pms/
1•awaxman11•43m ago•0 comments

AI: Brainrot Inducer or Cognitive Multiplier?

https://www.cjroth.com/blog/2026-02-12-brainrot
1•thoughtfulchris•44m ago•0 comments

Deft – a class and interface system for Clojure[video]

https://www.youtube.com/watch?v=dlW6YzwUZ-M
1•sammy0910•44m ago•0 comments

AI and consciousness: from objective descriptions to 'level zero'

https://randomseed.io/txt/ai-and-consciousness/
1•siefca•46m ago•1 comments

Cloudflare adds real-time Markdown rendering for AI agents

https://blog.cloudflare.com/markdown-for-agents/
5•thestackfox•47m ago•2 comments

A Read-Only Philosophical Archive on Restraint and AI Ethics

https://coexilia.io/coexilian-documents/
1•aegissolis•48m ago•1 comments

RFK Jr. food pyramid site links to Grok, which says you shouldn't trust RFK Jr

https://arstechnica.com/health/2026/02/rfk-jr-food-pyramid-site-links-to-grok-which-says-you-shou...
3•doener•48m ago•2 comments

Skip the Tips: A game to select "No Tip" but dark patterns try to stop you

https://skipthe.tips/
10•randycupertino•48m ago•4 comments