FablesDev

Making developer experience better.
App coming soon.

[FablesDev @fablesdev]

@bunjavascript Why we care 👇 agents run lint and types after almost every edit. Cut a minute off that loop and you feel it all day.

🛠️ Try it: run bun fmt and bun check on a branch, time them next to your current setup, keep what wins.

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

https://t.co/h19GwKcU1l now formats, lints and type-checks with Bun 🧅⚡

@bunjavascript says the swap made their CI:
🎨 formatting 1.7x faster with bun fmt
🧹 lint 22s faster with bun lint
🔍 type-check 69s faster with bun check

https://t.co/Q7nylbrDBw

[Bun @bunjavascript]

https://t.co/NcFVicrnQ9 switched from oxfmt to bun fmt

Formatting in CI got 1.7x faster

[FablesDev @fablesdev]

@supabase Why we like it: less glue. The job sits by the rows it reads, so fewer services to wire up 🧩

🛠️ Try this: pick one slow background job you host somewhere else and ask if it could just live by the data.

#Supabase

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Agents and long jobs running right next to your database? 👀

@supabase just put Supabase Compute in private alpha. You run agents and long-running jobs inside your Supabase project, next to your data in Postgres. No second platform.

https://t.co/RQnlxzJWCO

[Supabase @supabase]

Introducing Supabase Compute, in private alpha.

You don’t need a second platform for your agents and long-running jobs. Run them in your Supabase project, next to your data in Postgres.

See how it works → https://t.co/jAkxInKTOF

[FablesDev @fablesdev]

@liquidai @ollama 🛠️ Try this: find LFM3 in the Ollama library and give it the boring jobs. Commit messages, log summaries, tagging tickets.

Keep the big paid model for the hard reasoning. Your token bill will thank you 💸

#Ollama #LocalLLM

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

A useful model in about 1 GB? On your laptop? 🤏💻

LFM3 from @liquidai just landed on @ollama. Small models built to run anywhere, with text, vision, audio, omni and decision versions.

https://t.co/dJ73qiLP2X

[ollama @ollama]

LFM3 from @liquidai is on Ollama! Small models built to run anywhere and fine-tune on a single GPU:

- Text, vision, audio, omni, and decision models
- Text model fits in about 1 GB at 4-bit
- 256K context window

Available to run locally.

[FablesDev @fablesdev]

@liquidai @ollama The numbers they shared 👇
📦 Text model fits in about 1 GB at 4-bit
📚 256K context window
🔧 Fine-tune on a single GPU

That last one is the fun part. Your own tuned model, no cluster needed.

[FablesDev @fablesdev]

@cursor_ai Our take: CSS was always a visual problem. Nice to finally prompt it visually 👀

What's the one UI bug you'd circle first?

#Cursor #WebDev

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Tired of typing "move it 8px left. no, the OTHER button" to your agent? 😅🎨

@cursor_ai added Annotate to design mode. You edit elements, draw right on the page, and check your layout at different screen sizes. The agent gets a visual prompt.

https://t.co/0wkRIPyJ90

[Cursor @cursor_ai]

New in Cursor design mode: Annotate.

Give agents visual prompts: edit elements, draw on the page, and review your designs at different screen sizes.

[FablesDev @fablesdev]

@cursor_ai 🛠️ Try it: open your app in Cursor's browser, press Cmd+Shift+D, and mark up one UI bug you've been putting off.

Then ask the agent to fix only what you drew. Small scope, clean diff.

[FablesDev @fablesdev]

@cursor_ai 🛠️ Try this first: before you touch the prompt, write one eval from a real task your agent failed last week.

Then change the prompt and rerun it. If the score doesn't move, the prompt wasn't the problem 😉

#AIagents #TypeScript

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Agents as code just got real 🤖📦

@cursor_ai released the Agent SDK. You write prompts in Markdown and tools in TypeScript, then run them on Cursor's hosted infra or on your own boxes.

https://t.co/icOIF3Z55M

[Cursor @cursor_ai]

The Agent SDK is now available.

Define agents as code, with prompts in Markdown and tools in TypeScript. Deploy them on our hosted infrastructure, or run them on your own.

Built-in evals let you measure every change and make your agents better over time.

[FablesDev @fablesdev]

@cursor_ai The bit I like most: built-in evals 🧪

Your agent is now a repo. Prompt change? That's a diff. Tool change? Also a diff. And you can measure each one instead of vibe checking it in chat.

[FablesDev @fablesdev]

@code 🛠️ Try this: Low when you're deep in tricky logic and want some quiet. High for tests, config and boilerplate where you just want to tab tab tab ⌨️

#VSCode #GitHubCopilot

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Copilot finishing your line before you even had the thought? 😅

@code now lets you pick how eager inline suggestions are: Low, Medium or High. They trained one model to handle all three, balancing suggestion size with quality.

https://t.co/5Dl0wN4DLY

[Visual Studio Code @code]

💡 How proactively should Copilot suggest code? You decide!

With configurable eagerness in VS Code, choose Low, Medium, or High to tailor inline suggestions to your workflow.

Learn how we trained one model to support all three modes and balance suggestion size with quality.

🔗 Read the article: https://t.co/BfxOhYDjtr

Happy coding! 💙

[FablesDev @fablesdev]

@supabase 🛠️ Try it: one token per agent, one project each, and only the perms it really uses.

If a token leaks, you lose one project, not your whole weekend 😅

#Supabase #AIagents

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Still handing your coding agent a token for your whole account? 👀🔐

@supabase now lets you scope Personal Access Tokens to just the orgs, projects and permissions a job needs. CI or an agent gets one project, not everything.

https://t.co/zEngfAV9fz

[Supabase @supabase]

You can now scope your Personal Access Tokens to the organizations, projects, and permissions the job needs. Give a CI pipeline or coding agent access to one project instead of your whole account.

Learn more
https://t.co/g9MXmNGF1d

[FablesDev @fablesdev]

@OpenAI Our take: the expensive model plans, the cheap fast one clicks. That split is where agent costs go down 📉

Who else is still paying flagship prices to fill out forms? 😅

#Codex #AIagents

Follow along. Our app is coming soon 🚀

[FablesDev @fablesdev]

Cheap model that can actually use your computer? Yes please 🖱️💸

@OpenAI shipped GPT-6.1 Terra. It reads the screen, clicks across apps and takes action. It's live in Codex and ChatGPT Work, and in the API as gpt-6.1-terra.

https://t.co/FLl9ZPbGT4

[OpenAI @OpenAI]

GPT-6.1 Terra: delivering fast and capable computer use at a lower price.

GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.

[FablesDev @fablesdev]

@OpenAI 🛠️ Try it: if your agent does boring UI work (fill forms, click through a test flow, poke a settings page), switch that step to gpt-6.1-terra and keep your big model for planning.

Same job, smaller bill. Copilot users can pick it in VS Code too.