* llama.cpp, github version instead of hardcoded version
* llama.cpp, check if model is specified, if yes, run it, if not, then download model
* Use entrypoint for custom llama.cpp invocation
* `llama.cpp` is just raw executable. This I think is our new pattern.
* To run chat use the entrypoint: `pkgx +brewkit -- run llama.cpp`
Co-authored-by: James Reynolds <magnsuviri@me.com>
Co-authored-by: Max Howell <mxcl@me.com>
* Initial packaging of asciinema
Signed-off-by: Jacob Torrey <jacob@thinkst.com>
* Removed duplicated deps in test
Signed-off-by: Jacob Torrey <jacob@thinkst.com>
* Rename to the canonical asciinema.org for the main package
Signed-off-by: Jacob Torrey <jacob@thinkst.com>
---------
Signed-off-by: Jacob Torrey <jacob@thinkst.com>
* add fluentci package
fix test
* add deno to dependencies and update build script
* fix(fluentci.io): set skip-patchelf to true
* fix(fluentci.io): use skip: string | string[] syntax
* fix test
* fix test more
---------
Co-authored-by: Jacob Heider <jacob@pkgx.dev>
This caused all git dev-envs to implicitly depend on pkgx which would then cause a potentially older pkgx than the system installed one to be in most working envs. lol
Longer term a better solution would be nice.
In general having git in most dev-envs sucks since its deps are so intrusive. Would be nice to statically build parts of it or something.
* add(crystal-lang)
* move to bootstrap
this always hurts
* oops
* linux needs explicit -z
* linux not detecting the compiler well
* Only shim build
* needs ncurses
* shim needs link flags for eval?
* need link flags for test
* maybe what eval needs is --static
* the docs are wrong
* this should do it, i think
* perhaps static linkage
* punt
---------
Co-authored-by: Jacob Heider <jacob@tea.xyz>
Co-authored-by: Jacob Heider <jacob@pkgx.dev>
