Your agents.md is a good place for high level facts, but if you have something that requires a lot of info to explain (ie: if there is a complex build process, testing patterns, things like that), loading up your agents.md for every request may be a bad idea. Offloading that information to a skill ensures it's only included in the context if you're actually using it.
How about linking to a separate docs file from the Readme, same as how you'd split separate topics into different files for humans? The context cost is low and as far as I can tell it's pretty much how Claude's "memory" feature works.
That splitting is basically what a skill is, with some instructions as to when to load it. Depending on the harness used it might be quite equivalent, but not sure how easy it follows links in the readme compared to skills (which is just a glorified name of a readme anyways)
Right, it's all just text file management. My point is that if I can add a few lines of description with links in my readme/agents/etc instead of manually including skills in the prompt, without any downside, I'd rather do that.
Something else I want to add on to be more specific is that I use skills to document tasks that my model doesn't know how to do out of the box. For example, there is a jira cli[0] that I use for interfacing with jira. If I just say something like "add an issue for X on jira", the model will have no idea how to interface with jira. I could add that the jira cli is installed on my system, but it's not popular enough for the model I use to just know how to use the CLI, and it will end up spending a lot of tokens guessing how to use it, failing, reading the help output, trying again, etc. Adding a skill for jira lets me capture how to use the cli, and makes prompts like the original usually work first try. If something still requires the model to iterate with the cli, I will ask the model why it failed originally, and ask it to update the skill file accordingly, to avoid the same failures in the future.
