frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Painless Swift development on NixOS using rootless containers

https://carette.xyz/posts/distrobox_and_swift/
1•LucidLynx•3m ago•0 comments

The feature in OxCaml that more languages should steal

https://theconsensus.dev/p/2026/06/27/the-feature-in-oxcaml-more-languages-should-steal.html
1•birdculture•4m ago•0 comments

A hash proves the bytes, not the source

https://collider.ee/blog/2026-06-28-1500_a_hash_proves_the_bytes_not_the_source/
1•mog_dev•4m ago•0 comments

The database that refused to die: How Postgres survived its own creators

https://www.theregister.com/databases/2026/06/22/the-database-that-refused-to-die-how-postgres-su...
2•thunderbong•5m ago•0 comments

China Resets AI Race

https://www.wsj.com/tech/ai/chinese-ai-anthropic-mythos-cybersecurity-574b02c2
1•joe_the_user•6m ago•0 comments

Monlite Simple Infrastructure for AI Agent

https://github.com/qataruts/monlite
1•emadjumaah•6m ago•0 comments

GitHub – librepods-org/librepods: AirPods liberated from Apple's ecosystem

https://github.com/librepods-org/librepods
2•rbanffy•6m ago•0 comments

Organism II, a luminous deep-sea creature made from pure geometry

https://sand-morph.up.railway.app/organism
1•echohive42•7m ago•0 comments

Meet Jack

https://github.com/mlangford75-lgtm/jack
1•Mlangford75•15m ago•0 comments

Smartphones with Popular Qualcomm Chip Share Private Information

https://www.nitrokey.com/news/2023/smartphones-popular-qualcomm-chip-secretly-share-private-infor...
2•airhangerf15•20m ago•1 comments

Historical memory prices 1960-2026

https://dam.stanford.edu/memory-prices.html
3•vga1•22m ago•0 comments

The Cost Yagni Was Never About – By Kent Beck

https://newsletter.kentbeck.com/p/the-cost-yagni-was-never-about
10•kiyanwang•23m ago•0 comments

The Great American Tech "Crackdown" Is Looking Like China's

https://x.com/compose/articles/edit/2070615531422433280
2•ceohockey60•24m ago•0 comments

Scaling Engineering: Ownership over Hiring

https://greenido.dev/2026/06/11/what-changes-at-20-50-and-200-engineers/
1•kiyanwang•26m ago•0 comments

Sun Ra's Full Lecture and Reading List from His 1971 UC Berkeley Course

https://www.openculture.com/2014/07/full-lecture-and-reading-list-from-sun-ras-1971-uc-berkeley-c...
2•mmooss•26m ago•0 comments

Basic Computer Literacy

https://excipio.tech/blog/basic-computer-literacy/
1•lvales•27m ago•0 comments

Agents Are the New Product's Interface

https://www.hopsworks.ai/post/agents-are-your-new-product-interface
2•LexSiga•30m ago•0 comments

The LLM shoggoth meme is weirder than you think

https://hedonicescalator.substack.com/p/the-llm-shoggoth-meme-is-weirder
3•HedonicEscal8r•32m ago•0 comments

Lost your crypto access code? Be wary, there's a scam for that too

https://www.theguardian.com/money/2026/jun/28/scam-watch-panic-thats-just-what-fraudsters-are-wai...
1•Brajeshwar•35m ago•0 comments

A Streptomyces megacluster encodes synergistic biotin-targeting antibiotics

https://www.nature.com/articles/s41586-026-10647-9
1•Lihh27•36m ago•0 comments

Ask HN: Is your journal hard cover or soft?

1•herodoturtle•36m ago•0 comments

The Lost Art of Leisure

https://www.theatlantic.com/newsletters/2026/06/leisure-productivity/687726/
1•shadow28•39m ago•0 comments

High Schoolers and AI

https://malerchen.com/high-schoolers-and-ai/
1•pokemyiout•41m ago•0 comments

France's heat this week exceeded a 2050 forecast in 19 of 34 locations

https://news.slashdot.org/story/26/06/27/2146244/frances-heat-this-week-was-worse-than-a-dire-sce...
4•logickkk1•42m ago•2 comments

Continuity-Enhancing Degree Elevation and Splits

https://graphics.cs.utah.edu/research/projects/continuity-enhancement/
1•E-Reverance•43m ago•0 comments

'Born to make people laugh': Comedy legend and Jewish icon Mel Brooks turns 100

https://www.timesofisrael.com/born-to-make-people-laugh-comedy-legend-and-jewish-icon-mel-brooks-...
1•sbuttgereit•44m ago•0 comments

Ask HN: Is WordPress the best way to create new websites for beginner

3•anitroves•52m ago•8 comments

AI and the Crisis of 'Classical Liberalism'

https://compactmag.substack.com/p/ai-and-the-crisis-of-classical-liberalism
1•paulpauper•54m ago•0 comments

The Tragic Birth of FM Radio (2006)

https://www.damninteresting.com/the-tragic-birth-of-fm-radio/
3•bookofjoe•54m ago•0 comments

Best Books of the 21st Century

https://www.nytimes.com/interactive/2024/books/best-books-21st-century.html
1•paulpauper•54m ago•0 comments
Open in hackernews

Show HN: Caliper – pass@k reliability testing for Claude Code and Codex skills

https://github.com/edonadei/caliper
2•edonadei•1h ago
Skills for Claude Code and Codex are hard to test. What I mean by hard is that there's no standard way to do it. You evaluate the skill once on something, it looks like it works. You publish it. Then the new super model releases (GLM 5.2 anyone?), it will quietly break for some part, and you won't find out until your users complain.

I also faced the same problem, so I tried to build something lightweight to stop doing that. Caliper.

It's a local and lightweight harness that runs a skill k times in isolated environments and gives you a pass@k score (How much times it succeeded in these k times). As a non-deterministic technology, you can't just say "it worked once". You need to answer how much it passed in k times.

You define success in a YAML spec. I picked YAML to keep a schema and make it still readable for a human. You either use a LLM judge, a Python assertion, or both:

Here's an simple evaluation example with a JSON extraction, so you write this in a YAML file:

  tasks:
    - name: Extracts action items as clean JSON
      prompt: "Read /tmp/transcript.txt and write the
               action items to /tmp/actions.json."
      expect: "A valid JSON array where every item has
               owner, task, due. No markdown fences."
      assert: |
        import json
        items = json.load(open("/tmp/actions.json"))
        assert isinstance(items, list)
        assert all({"owner","task","due"} <= i.keys()
                   for i in items)
Then with the CLI, you'll run it:

caliper run extract-actions.eval.yaml --k 5 --baseline

What's cool about the --baseline flag is that it will re-runs everything without the skill, so you can see whether the skill is doing the work or the base agent was going to pass anyway:

  ID      Task                           k(5)  pass@k
  task-1  Extracts action items as JSON  5/5   100%  PASS
  With skill   100%
  No skill      60%
  Delta        +40%
Most models know how to get the JSON right most of the time (JSON extraction was solved by 2 years old already). But that's it, "most of the time" is the bug. That delta shows how the skill actually helped. (It's sometimes 0%, sometimes -100%!)

I also created two skills you can get started right away with your favorite harness, e.g. Claude Code, Codex or Pi:

- evaluate-skill: run and manage evals without leaving your workflow

- grill-skill: reads your SKILL.md, interviews you about what "good" looks like, writes a 3-task spec (happy path, edge case, adversarial), and runs it

You can install the skill with the command: npx skills@latest add edonadei/caliper

I for now support claude-code, codex, pi, claude-api, openai-api. You can run the agent and the judge as separate backends, so you can run a skill on one and judge with another.

GitHub: https://github.com/edonadei/caliper PyPI: https://pypi.org/project/caliper-eval/

Of course, it's a first step. I think the autorater layer can be vastly improved, more handholding to create and iterate on evaluation specs, supporting more harness, why not including this layer into a self-improvement bigger system?

If you're also building agentic evaluations, I'm genuinely interested to hear how you are handling that.

Comments

edonadei•25m ago
btw FYI, a really good article on evaluation, I vastly based my research and iteration from it https://www.anthropic.com/engineering/demystifying-evals-for...