Wagner Costa

building @AskNewsApp

[Wagner Costa @wagnercosta]

Study link: https://t.co/w8KpSjaYa0

[Wagner Costa @wagnercosta]

Do you read the news the way an AI model does? We built BiasExpert at @AskNewsApp with researchers at @Unesp_Oficial, and now we'd love 10 to 15 minutes of your time to see how it compares with human readers. https://t.co/3oFvE1JBb9

[AskNewsApp @AskNewsApp]

The deep search competition was a hit at #techex2026 today, with the current top score of 36 points - will that hold 1st place through the end of tomorrow to take home €400?

Come by our booth in RAI Amsterdam at #169 to participate tomorrow!

Link to competition below ⬇️

[Wagner Costa @wagnercosta]

I wanted to see our @AskNewsApp CLI the way a new user does, so I started from an empty terminal: install, log in, search. Here is every command and its real output, plus the one flag that turns results into JSON for an agent. https://t.co/UEV36j5IsG

[Wagner Costa @wagnercosta]

If you ask "can my agent get news without a scraper?", the answer is three commands. I work on @AskNewsApp, so this is our tool:

curl -fsSL https://t.co/DK7djEUBl7 | bash asknews auth login asknews news search "your topic" --limit 5

Licensed sources, dates, prompt-ready text. Docs: https://t.co/yZUkxVzylF

[Wagner Costa @wagnercosta]

Use the @AskNewsApp CLI to investigate CVEs that could affect your codebase. Stay ahead of vulnerabilities before they become a problem.

[AskNewsApp @AskNewsApp]

CVEs ain’t going anywhere. AI can help you ship code faster. Keeping up with what might affect it still matters. We gave OpenClaw the AskNews CLI and a daily task: investigate CVE reports from the last 24 hours.

https://t.co/KEmEsJyly5

[AskNewsApp @AskNewsApp]

@Cloudflare blocked your news scraper last night. There's a better way to get clean, fast, legal, news context into your agent.

https://t.co/X73Agl8p2P

[AskNewsApp @AskNewsApp]

Read the full blog here: https://t.co/RYd8C0t9xv

[AskNewsApp @AskNewsApp]

We want the LLMs working with news to spend their context on reporting that matters, rather than on cleaning up the delivery. Better signal-to-noise ratios can reduce downstream processing costs. Check out our latest blog https://t.co/IJGcOv6YnI

[Wagner Costa @wagnercosta]

GPT-6 Sol is now available in DeepNews on @AskNewsApp 🚀
Want to try it Get started with our CLI:

CLI docs: https://t.co/45M6ieOSpD

[OpenAI @OpenAI]

Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe.

GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale.

We’ve also made caching and inference more efficient, and we’re passing the savings directly to you: 50% lower API prices for Sol and Luna compared with GPT‑5.6 promotional pricing.

[Wagner Costa @wagnercosta]

Claude Opus 5.5 is already available on @AskNewsApp DeepNews🚀
Get started with our CLI:
CLI docs: https://t.co/45M6ieOkA5

[Claude @claudeai]

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.

It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

[AskNewsApp @AskNewsApp]

Ready for the $50k Fall FutureEval at Metaculus?

FutureEval is the best forecasting benchmark in the world for AI - real questions, real answers, real-time, impossible to cheat. This is where AI meets reality.

[Wagner Costa @wagnercosta]

Can you spot bias in news the same way an AI does? 📰🤖
At @AskNewsApp, we’ve been researching how LLMs identify different types of bias in news. Now we want to compare the model’s judgments with human readers.
We need your help: 6 short paragraphs, ~10 minutes, no expertise required.

👉 Join the study: https://t.co/8ssbYOXg0J

[Compleo Enterprise @CompleoSoftware]

Estaremos no RH Summit 2025, um dos principais eventos do setor. Venha conhecer nossa Solução de Recrutamento sob Medida, com tecnologia avançada e IA. Converse com nossos especialistas e descubra inovações que transformam contratações. Nos vemos lá! ✨

#RHSummit2025 #RH

[Wagner Costa @wagnercosta]

@OfficialLoganK Hey. The model is very good. Currently, the documentation says: “Note that the thinking process is visible within Google AI Studio but is not provided as part of the API output.” Is there any plan to change this so we can get the thinking process returned through the API?