Introducing sshrunner: Parallel SSH for DevOps and Automation

Stop memorizing IP addresses. 🛑 I got tired of typing ssh root@192.1xx.xx.xxx from memory every time I needed to check a server. So, I built ssh-to-server — a tiny CLI tool that turns your SSH config into an interactive menu. Why use it? 📂 Zero Config: It automatically reads your existing ~/.ssh/config. ⌨️ Interactive UI: Pick a server using arrow keys or quick number shortcuts. 🛠️ Respects your setup: Works with your IdentityFile, custom ports, and proxy jumps. Is it revolutionary? No. Does it save me from 20 daily typos? Absolutely🤓 Built with Python, Click, and a healthy dose of terminal customization procrastination. 📦 Install it: uv tool install ssh-to-server (or pip install ssh-to-server) 🔗 GitHub: https://lnkd.in/eh--kjfZ If your SSH config is longer than your shopping list, this one's for you. 🤝 #Python #DevOps #CLI #OpenSource #Productivity

Your GitHub Actions are pinned. But are they actually secure? The Trivy compromise exposed a blind spot for me. Here were workflows doing everything right: SHA-pinned references, locked dependencies. And the action itself was pulling in unversioned artifacts at runtime. The pinning was theater. SHA-pinning protects the code at checkout. But a pinned action can still curl down releases/latest, run go install @latest, or FROM :latest in a Dockerfile. Nothing in your workflow file tells you that's happening. My friend William Woodruff's zizmor (https://zizmor.sh) is the tool I'd point anyone to for the static side: template injection, excessive permissions, credential leaks in your workflow files. If you're not running it, start there. But zizmor is intentionally scoped to static analysis of workflow files. It's not looking at what actions do when they actually run. That's the gap I'm trying to close with pinprick. It fetches action source code and scans for unversioned runtime fetches: pipe-to-shell patterns, unpinned package installs, unversioned API calls, FROM :latest in Dockerfiles. The behaviors that bypass everything you've pinned. It also handles resolving tags to full SHAs and checking pinned actions for newer releases. Outputs SARIF for GitHub code scanning. Ships with pre-audited results for popular actions so you get signal immediately. Written in Rust. Open source (AGPL-3.0). https://pinprick.rs #github #opensource #rust #devsecops #supplychainsecurity

𝗛𝘁𝘁𝗽-𝗦𝗲𝗿𝘃𝗲𝗿 𝘃𝟱 𝗶𝘀 𝗹𝗶𝘃𝗲!!! A few months ago, I developed a C++ library for creating HTTP servers that run directly on the machine, unlike Express.js or FastAPI, which run on Node.js and Python respectively. It worked fine, but there were some fundamental issues that made it unusable for most cases. In the latest version (v5.x.x), these problems have been addressed. Changelog: Support for 𝗪𝗶𝗻𝗱𝗼𝘄𝘀 𝗢𝗦. Stream based request and response bodies. 𝗠𝘂𝗹𝘁𝗶𝘁𝗵𝗿𝗲𝗮𝗱𝗲𝗱 servers capable of handling multiple requests concurrently. Shared response sending thread. Separate event manager for reading data and sending data. Better handling of malformed requests. Consistent API. Bug fixes. You can checkout the repository on GitHub. Contributions and suggestions are welcome. GitHub: https://lnkd.in/gj6-pH4J Latest Release: https://lnkd.in/gpWVFrqz #Network #lowleveldevelopment #Cprogramming #cplusplus #ComputerNetwork #SocketProgramming

Your Dockerfile's FROM line is a trust decision. 🔐 A developer using Label Studio just opened a GitHub issue asking the team to swap their python base image for Docker Hardened Images (DHI) after repeatedly hitting vulnerabilities in image scans. It's a pattern I see constantly: → Team builds a great app → Base image carries hidden CVEs → Security scan fails → Everyone scrambles The fix? Start from a hardened base. DHI ships with a minimal attack surface, no unnecessary packages, and regular security updates baked in. And since earlier this year, it's free under Apache 2.0. Small Dockerfile change. Big security posture improvement. 🔗 See the issue: https://lnkd.in/gQ8YNnJx #ContainerSecurity #Docker #DHI #DevSecOps #DockerHardenedImages #Kubernetes

“App started breaking under the first production load test, even though the structure was correct.” Sometimes it’s not just about having a perfect application structure but about proper configuration tuning. That’s why I’m sharing another comparison, the earlier run without tuning (present on my profile) and the optimized version after tuning. At ~300 concurrent users, latency behavior degraded sharply, P99 reached ~2.7s and max spikes went up to ~7s, even though average response time still looked normal. The issue wasn’t the API structure or queries; it was how DB connections were being handled under concurrency. Pooling wasn’t properly enforced, so requests started queuing instead of executing in parallel. After fixing proper pool acquisition and ensuring each request correctly borrows/releases a connection, performance stabilized under the same load. Same system. Different behavior under pressure. #fastapi #python #backenddevelopment #postgresql

Finally figured out MCP. Turned a couple of local Python scripts into MCP servers and wired them into Gemini CLI as skills. The value? No more memorizing syntax, Gemini CLI handle it. Real value? For now, it’s more about understanding the technology than using it. I think it’s a great technology, but I’m still figuring out where it’s actually valuable in my current projects. Also, now I see how to build MCP as an API wrapper or run it as a web service.

There’s a quiet risk sitting in a lot of networks today, configuration backups. You can spend your day logging into routers one by one… or write a script that does it for you while you focus on more important work. A lot of teams don’t notice the gap in their backup process until something breaks. Configuration backups sound simple, but in many environments they’re still manual, inconsistent, and easy to forget. And when a core switch goes down? That’s when the problem shows up. The goal is simple: move from reactive firefighting to controlled operations. Instead of logging in manually every time, a script handles the SSH connections using Netmiko. Backups are automatically saved with timestamps, so you can actually track what changed and when. More importantly, when something fails, your MTTR (Mean Time to Repair) drops significantly, because the configs you need are already there, organized and ready. This is one of those quick wins that makes an immediate difference. I shared a step-by-step guide in the Python for Network Automation Beginner series: https://lnkd.in/eyVkppf2 #DevNet #NetworkAutomation #CiscoInsiderChampion #NetOps #Python #Scontinum #Learningisacontinum

🚀 InstapaperScraper v1.3.1rc1 Pre‑Release Try it out today: 👨💻 pip install instapaper-scraper==1.3.1rc1 I’ve just published a new release candidate on PyPI and GitHub, continuing to refine stability and security ahead of the v1.3.1 stable rollout. Highlights: 🔧 Compatibility update – Dropped Python 3.9 (EOL) support. 🔒 Security fix – Prevent path traversal in output filenames. 🔒 API validation – Ensure folder_id and slug values are URL‑safe. ⬆️ Dependency bumps – Updated cryptography, urllib3, and requests to latest versions. 📌 Dev environment – Version numbers added to dev dependencies for consistency. 💬 Share feedback and issues on GitHub to help shape the stable release. #InstapaperScraper #Python #OpenSource #ProductivityTools

Do you keep fighting with CVEs reported by scanners like Trivy, many of them 𝗳𝗮𝗹𝘀𝗲 𝗽𝗼𝘀𝗶𝘁𝗶𝘃𝗲𝘀? Try using 𝗺𝗶𝗻𝗶𝗺𝗮𝗹 𝗰𝗼𝗻𝘁𝗮𝗶𝗻𝗲𝗿 𝗶𝗺𝗮𝗴𝗲𝘀! For those of you who tried, you probably found that searching for minimal images is cumbersome. I wrote a blog post that significantly reduces your research time. It explains the pros and cons of minimal images, and provides a list of free and paid minimal images for PHP, Python, Java, C#, Node.js, and bare Linuxes (into which you would copy binaries compiled with C/C++, Go, or Rust). I originally wrote that post in October last year, and 𝗜 𝗷𝘂𝘀𝘁 𝗿𝗲𝗳𝗿𝗲𝘀𝗵𝗲𝗱 𝗶𝘁 𝘁𝗼 𝗽𝗿𝗼𝘃𝗶𝗱𝗲 𝘆𝗼𝘂 𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝗻𝗲𝘄𝗲𝘀𝘁 𝗶𝗺𝗮𝗴𝗲𝘀 𝗮𝗻𝗱 𝘃𝗲𝗿𝘀𝗶𝗼𝗻𝘀 (and added Docker Hardened Images / DHI). Find the link in the comments. I hope it helps you. ---- Some background about minimal images, if you lack context: Minimal images omit components like shells or package managers, which are typically not needed at runtime. This significantly reduces the attack surface, thereby enhancing security. Because the fewer components that exist in an image, the fewer potential vulnerabilities there are. Scanners report far fewer CVEs for minimal images, reducing the frustration and unnecessary workload for your teams to analyze CVEs. #docker #container #image #security #comparison #supplychain #distroless