I've continued to use this as my primary terminal "window manager" and wanted to share in case others might find it useful.
I would love to hear about other's terminal-based workflows and any other tools you may use with similar functionality.
We mostly work on backend and workflow tooling, but we needed a way to turn Figma designs into frontend code as a kickstart for prototyping. It takes a Figma frame and converts it into React + Tailwind components (plus assets).
If you want to try it: You can run it locally or use it via the VibeFlow UI to poke at it without setup (https://app.vibeflow.ai/)
Features:
- Real-time NWS alerts with safety tips - Snow/ice/precip accumulation forecasts (+wind) - Dynamic preparation checklists based on your alerts - Supply calculator for your household size - Regional weather news
It's free, no login required, works on any device. Just enter your ZIP.
https://jeisey.github.io/stormwatch/
Uses NWS and GDELT APIs and open source. Feel free to fork and modify however you'd like.
For builders: - Used an API-testing agent to verify all endpoints, response patterns, types, and rate limits - Used a scope & validation agent to keep the slices simple, focused, and tested - VS-code Copilot (Sonnet 4 for dev agents + Opus 4.5 for scope and validation)
Examples
Simple function:
from polymcp.polymcp_toolkit import expose_tools_http
def add(a: int, b: int) -> int: """Add two numbers""" return a + b
app = expose_tools_http([add], title="Math Tools")
Run with:
uvicorn server_mcp:app --reload
Now add is exposed via MCP and can be called directly by AI agents.
API function:
import requests from polymcp.polymcp_toolkit import expose_tools_http
def get_weather(city: str): """Return current weather data for a city""" response = requests.get(f"https://api.weatherapi.com/v1/current.json?q={city}") return response.json()
app = expose_tools_http([get_weather], title="Weather Tools")
AI agents can call get_weather("London") to get real-time weather data instantly.
Business workflow function:
import pandas as pd from polymcp.polymcp_toolkit import expose_tools_http
def calculate_commissions(sales_data: list[dict]): """Calculate sales commissions from sales data""" df = pd.DataFrame(sales_data) df["commission"] = df["sales_amount"] * 0.05 return df.to_dict(orient="records")
app = expose_tools_http([calculate_commissions], title="Business Tools")
AI agents can now generate commission reports automatically.
Why it matters for companies • Reuse existing code immediately: legacy scripts, internal libraries, APIs. • Automate complex workflows: AI can orchestrate multiple tools reliably. • Plug-and-play: multiple Python functions exposed on the same MCP server. • Reduce development time: no custom wrappers or middleware needed. • Built-in reliability: input/output validation and error handling included.
Polymcp makes Python functions immediately usable by AI agents, standardizing integration across enterprise software.
It focuses on FFT, filters, PSD, STFT, DCT and Android compatibility, aiming to fill the gap for DSP-heavy workloads on JVM and Android.
I run safety-critical math. "Close enough" doesn't cut it.
I ran the exact same workload on my local M1 and my H100. The numbers didn't match. Standard libraries (NumPy, PyTorch) drift because standard GPU schedulers fuse and reorder ops differently than CPUs.
I built my own engine (LuxiEdge) to fix it. Now I get bit-exact matches across both architectures. 0.00% drift.
Here is the repo and the reference hash. If you think your standard tools can do this, go ahead and try to match it. You can't.
Claude Code has proven that coding agents work best when they are local, bringing developers back to the terminal.
Terminal UIs are fast and great with shortcuts, shell mode, and developer-friendly workflows. But they are limited in history and image display, and the experience varies by terminal and OS. Many of them flicker (buuuut not AdaL CLI ).
Most importantly, they can be quite intimidating for non-technical users.
This led us to explore new possibilities for a coding agent interface. What if you could get the best of both worlds: - the same core local agent that does tasks exactly like AdaL CLI - combined with a web UI with no limits on UI/UX
This can be especially powerful for design-heavy and more visual workflows
Available at: https://sylph.ai/
As a developer, I’ve always struggled to remember every command and its specific options. Whenever I need to use tools like Docker, Git, or psql, I find myself leaving the terminal to check documentation or scrolling through verbose --help text just to recall a specific flag. I usually know what I want to do, but I forget the exact syntax. I didn't want to waste time switching to Google or ChatGPT just to find a one-line command.
I developed Ask CLI to solve this. It has been a game-changer for my workflow. Now, when I forget a command, I simply ask my terminal. It gives me a fast, precise answer—exactly what I need—without breaking my flow.
It is incredibly easy to use: just select an AI model, set your API key, and start chatting naturally with your terminal.
Examples:
$ ask how to run a docker container with env variables
$ ask how to setup my local git account
You can also use the "what" and "how" aliases for a more natural feel:
$ what is chmod
$ how to print all the env variables
You can use Ask CLI with popular hosted models (Gemini, Claude, ChatGPT) or with local models and external providers that support OpenAI-compatible APIs (Ollama, llama.cpp, LM Studio, etc.).
Ask CLI is free and open-source. Check it out here:
I built a tiny native menu bar app I'm calling Pingaroo. It separates router latency from internet latency and shows real-time signal quality/noise graphs to help diagnose intermittent lag. Written in 100% Swift/SwiftUI and fully open source.
Pingaroo is a recreation of the WhyFi app by James Potter based on screenshots, with a few personal riffs and improvements.
Feedback welcome!
It automatically generates short links and uploads to any S3-compatible storage you own.
Here is the link: https://gofwd.to/screenfwd
Try it out, all feedback is welcome :)
Well let's be clear on what you actually miss. No one misses that feeling of having to go and be there 8 hours. But many people miss friends. They miss being part of a crew. Going to lunch, hearing about other people's lives in person not over zoom.
Join a co-working space you say? Yes. We have. It's like walking into a library and trying to talk to random people and getting nothing back. Zero part of a crew feeling.
This app helps you find a crew and meet up for work and get that crew feeling.
This is my first time using cloudflare workers for a webapp. The free plan is amazing! You get so much compare to anything else out there in terms of limits. The sqlite database they give you is just fine, I don't miss psql.
I’ve been working on a small project called Kontra and just released it.
Kontra is a data quality measurement engine. You define rules in YAML or Python, run them against Parquet, CSV, or database tables, and get back violation counts and sampled failing rows.
The main goal was to avoid doing more work than necessary. Instead of treating all rules the same, Kontra separates execution paths. Some checks can be answered from Parquet metadata alone, others are pushed down to SQL, and full in-memory scans only happen for rules that actually need them. The guarantees differ, and Kontra is explicit about that rather than hiding it.
Under the hood it uses DuckDB for SQL pushdown on files and Polars for in-memory execution. It also supports profiling datasets, drafting starter rules from observed data, and diffing validation runs over time. Rules can carry user-defined context, and runs can be annotated after execution without affecting validation behavior.
It works as both a CLI and a Python library.
Happy to answer questions or get feedback.
Built a simple tracker focused on cholesterol. Log meals, see lipid breakdown, track trends. I believe snacks and sugar were my main issue.
Stack: Angular 17 + NestJS + Supabase
Started January 1st, already lost 3kg. Same breakfast daily (psyllium, oats, chia, skyr, whey, berries), cut sugar from daily to once per week.
Free during beta. Looking for feedback on whether strict diet cutting or 80/20 approach is more sustainable long-term.
The main purpose is to scratch my own itch when copying large files onto a USB drive. I want to be able to see the progress of sync (or "eject") operations whose feedback usually amounts to "wait" or "finished" !
At the CLI, of course, you can instead run progress [1] or just manually grep for "Writeback" in /proc/meminfo [2] but personally I like having something graphical!
[0] https://github.com/dcminter/meminfo-rs
[1] https://github.com/Xfennec/progress
[2] https://man7.org/linux/man-pages/man5/proc_meminfo.5.html
We have a Slack group with 30+ people and introduced this agent as a way for non-engs to push simple PRs and changes. This saves engs valuable time and context switching costs.
After a while, we expanded to this being a full-blown agent and you can see some interesting use cases in the blog post: https://blog.kilo.ai/p/announcing-kilo-for-slack
Any feedback is welcome!