diff --git a/.github/assets/boost-how-to-use-dark.gif b/.github/assets/boost-how-to-use-dark.gif new file mode 100644 index 0000000..ce35395 Binary files /dev/null and b/.github/assets/boost-how-to-use-dark.gif differ diff --git a/.github/assets/boost-how-to-use-light.gif b/.github/assets/boost-how-to-use-light.gif new file mode 100644 index 0000000..7142806 Binary files /dev/null and b/.github/assets/boost-how-to-use-light.gif differ diff --git a/README.md b/README.md index 4cecccc..de8f0a0 100644 --- a/README.md +++ b/README.md @@ -49,6 +49,14 @@ Boost never trades quality for savings. It trims only what's safe to drop, so agent output stays just as sharp. Our [Terminal-Bench 2.0 benchmark](https://boost.jfrog.com/blog/benchmarks-terminal-bench/) shows it: identical task pass rate, ~12% lower cost — Boost keeps agents optimized without ever breaking their stride. +
+
+