Skills and MCP are useful for when you need to repeat specific processes, perpetually. Without them the task description and reasoning falls out of the context window, or is compressed, and the process fails.
An agent will eventually forget, or hallucinate, guardrails and requirements. Yes to AGENTS.md, but when you're actively managing the whole context window in a long-running task you don't want to just keep jamming stuff in there and hope for the best. Skills help budget tokens and stabilize around specific outcomes.
If your use case is not agentic, as you build a skill corpus you can begin having the model reason at higher and higher levels about the outcomes you're aiming at.
Eg: I'm super lazy now and ask Claude to launch the project instead of just running the command myself. This is probably best done as a skill.
Me too just use AGENTS.md and it seems to work. I don't understand what problem MCP is trying to solve and skills just sounds like something you can do in AGENTS.md
I would guess the top 10% of actual performers do the same - the people who talk about harnesses and chain multiple systems together etc will be mid table somewhere
Just getting into homebrewing / jailbreaking old handhelds and consoles, and I really do feel like I missed a golden age of hacking that I would have loved. That said, I don't know if the tenacity I had back then would be enough to endure the steeper learning curve. Still, I would have loved to see my own work running on my DS.
I don't have an issue with the style personally. I actually like being able to navigate the whole page without clicking through pages. Looks like the content is the same too? Although that's just a glance comparing random snippets.
Despite the timestamp, I originally uploaded this a day ago and can no longer edit the title. My bad! I didn’t notice how dated the page was until it was too late. Thanks for the link, I’ve been mulling over getting a DSPico to tinker on my old DS.
Seeking a junior/entry/mid role at a startup or contract work. I interned at Google and have worked with a YC-backed startup. I can do full stack and have interests in systems engineering, distributed systems, and AI engineering.
Most recently, I built an operating system that plays Pong (https://github.com/Daves1245/horizonOS). I'm currently working on implementing the transformer architecture from scratch, and plan to make my own inference engine as well. My other projects include a multiplayer leetcode site (https://ternarysearch.com) inspired by binarysearch.com, a discord bot to answer questions about a github repo using RAG, a shabby discord clone to learn system design, a community-driven site that vibe codes itself, and home labbing.
reply