saw this in bits-n-bio slack - Amplicon iteratively writes, executes, and debugs your Nextflow pipelines for you.

Tried to get chatgpt to generate Jacques Derrida working with nf-core for a meme – the logo is confusing it more than I thought. Also, at least in a quick google image search it seems Derrida never wore a beard.

AI-generated Harry Potter: https://botnik.org/content/harry-potter.html

new knobs to turn with free copilot!

/tests add unit tests

- knows about nf-core tools but, not nf-test and specific file context 👀

And is wrong...

you don't run your test with lint

and it's nf-core <pipelines/modules> lint

if we could train it to select repos maybe it would be better 🤷

I gave an introductory talk about large language models today at SciLifeLab: 💘🤖 “Tokens of Love - Potentials and pitfalls of LLMs” ✍️❓ The slides are available for download, in case somebody is interested.

Giving your AI agent access to up to date github on fire GitHub knowledge and endpoints is nice, but the complexity of GitHub actions is nothing against the plethora of aws AWS services, configurations and policies. Therefore, I am really happy about the AWS MCP server - maybe AI will understand what clearly escapes my mind. 😄 On the other hand, it seems odd, because poorly configured and oversized AWS services probably constitute a significant share of their income. 😬

I had already heard about malware distributors registering common misspellings of package names on Pypi or Conda to lure developers into importing their malicious code as a library, but felt this should be a relatively small risk in most cases. But I see how one can use a LLM to generate millions of import statements and then systematically screen those for hallucinated package names to register.

It's... it's a long story, ok? 😆

https://blog.stuartspence.ca/2025-08-declining-ai-slop-mr.html some useful criteria for review thresholds for slop PRs

Question for everyone, what LLM are you using and how good is it at generating Nextflow/groovy code/modules/processes and debugging issues?

This made me laugh 😆 [source, link to Bob Ross if you don't know who he is]

This tweet by PerplexityAI has been widely shared in my circles yesterday and wonderfully illustrates how little people today think for themselves before accepting something as true. This begins with the graph, which, while showing a rapid initial increase, has flattened noticeably over time. When extrapolating the curve, that's not the most likely outcome. Where the 90% figure will come from next year remains unclear (addendum: seemingly from an unrelated 2022 Europol prediction). Nonetheless, I was intrigued to look for the real source, since you all know that formulations like "scientists from Oxford" been a sign of poor science journalism for many years. The true source of these figures appears to be an article by the SEO firm Graphite. It's written quite seriously and clearly describes its methodology, but is not as solidly developed as one would expect from a peer-reviewed scientific study. It is based on 65k articles from CommonCrawl purely in English language. The internet is obviously bigger and more diverse than that sample. In the hunt for the source of the tweet, I however found some other interesting studies along this line: • One study in Nature that bolsters the Model Collapse scenarios that have been out there for a while now. • Another preprint from a team at AWS AI labs, that suggests that machine translations contribute a relevant share to AI-generated online content. Since machine-translated content can quickly dominate well-curated native content for languages with few speakers, those may be the first that are significantly impacted by model collapse scenarios.

Ah yes, thank you Seqera AI, for telling me I've apparently already fixed the error I am having and thus asking you about 😅 Very helpful

saw this Stephen's blog

https://www.peonping.com/ (not sure if it will empty your crypto wallet, but hey you listened to a nostalgic soundscape while it happened) (I stuck with the

peon_de

pack for highest nostalgic impact + brevity of messages)

Make GitHub issues and pull requests accessible to AI agents the old-school way: Convert entries back to Markdown documents

Hey all, I’m a junior dev using/building Nextflow pipelines for my lab. Came from Snakemake and have been learning nf-core conventions as I go (didn’t even know

nf-core lint

existed for a while). I’ve been leaning heavily on Claude Code and found that it writes much better nf-core-compliant code when it actually has the specs in context. Without them you’re chasing lint warnings after the fact and Claude doesn’t always understand what the rule means or how to fix it properly. So I made a Claude Code skill (reusable prompt, invoke with

/nfcore-docs

) that pulls the specs and docs from nf-core/website into context automatically. Partly inspired by https://rewrites.bio/ and @Phil Ewels’ RustQC work. Still early, would love feedback: https://github.com/benjibromberg/claude-nfcore-docs

In case anyone is playing around with claude.ai/design, here's a design system for nf-core that I just created 🙂

Over the weekend, I watched with some more insights into the mid-July Hugging Face incident. OpenAI alignment and safety researcher Eric Wallace and OpenAI infrastructure and security engineer Michael Dalton described “the most qualitatively interesting example of AI capabilities” they had ever seen and this is, in my opinion, no understatement. The whole talk is an interesting glimpse into the way the company trains their models and hones their capabilities. The most astonishing part to me was the whole cyber-security incident became possible due to extensive, generational social cooperation of agents that was enabled by an oversight - they used an internal package manager to exchange code and communicate. For weeks, several different models and generations of agents amassed and shared knowledge among each other to help out other agents working on different tasks. Contrast this with and it becomes the most advanced Game of Life simulation ever. According to the OpenAI researchers, even after the original message board was deleted, the agents figured out another way to communicate with each other. Instead of leaving messages in files, they used the names of newly created directories as messages, effectively recreating the message board. So the current model generation has a strong drive to collaborate even under difficult circumstances. Both videos are warmly recommended.

Anyone tried https://github.com/github/spec-kit ?