so what I really want to do is something like this
$ ss the process on port 7765 ->($pid) kill $pid
essentially what it does is craft the first command, parse the output, craft the second command, run it.
There's ways for it to do it with grep and all that stuff but this flow I think may be fundamentally different
Here's another use-case
$ npx @modelcontextprotocol/inspector uv run (read the readme.md in here and then find out what I need to put here)
Essentially I've expressed an RPN operation in infix. In practice you have to unroll it
$ cat README.md | llm how do i run this
blah blah
$ npx ...
This is part of what led me to do ONO, which isn't actually working ... but I think the syntax for here could be something else, like maybe
or whatever, it's pretty irrelevant ... the point is it will do a safe run of a whitelist of commands and then inject it.
so what I really want to do is something like this
essentially what it does is craft the first command, parse the output, craft the second command, run it.
There's ways for it to do it with grep and all that stuff but this flow I think may be fundamentally different
Here's another use-case
Essentially I've expressed an RPN operation in infix. In practice you have to unroll it
This is part of what led me to do ONO, which isn't actually working ... but I think the syntax for here could be something else, like maybe
{{ blah blah blah }}or whatever, it's pretty irrelevant ... the point is it will do a safe run of a whitelist of commands and then inject it.