Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Bump vite-task git dependency to the latest main commit. Use when you need to update the vite-task git-dependency crates (vite_task, fspy, pty_terminal_test, and friends; the authoritative set lives in Cargo.toml) in vite-plus.
Run the standard vite-plus release process end-to-end as the release manager. Covers preparing the release PR, syncing the NAPI binding version, writing the categorized changelog, smoke-testing via a preview build, handling release-branch CI failures, merging, and post-release verification and announcements. Use when asked to cut, prepare, or manage a vite-plus release (e.g. "release v0.2.3", "act as release manager").
Compare tsdown CLI options with vp pack and sync any new or removed options. Use when tsdown is upgraded or when you need to check for CLI option drift between tsdown and vp pack.
Drive and capture vp's interactive (clack) prompts in a tmux session to manually verify interactive UX and catch spinner-over-prompt bugs. Use when asked to test/verify/capture an interactive vp command's prompts (vp migrate, vp create, ...), reproduce a prompt-rendering bug, or show the real interactive CLI output for a PR.
Guidelines for kernel correctness testing and coverage in fla/ops/** and related modules, including common Triton grid/addressing pitfalls. Helps decide what tests to add or run before an MR.
Workflow for FLA backend dispatch decorators and backend implementations. Use when touching fla.ops.backends, @dispatch-decorated functions, BaseBackend subclasses, backend verifier methods, backend env vars, or backend tests.
Checklist and workflow for preparing an MR/PR in the FLA repo. Covers CONTRIBUTING.md compliance, test plan, benchmark evidence, and PR body structure.
Guidelines for NVIDIA GPU kernel / Triton / Gluon / TileLang / CUDA backend performance work in the FLA repo. Covers profiling workflow, hardware baselines, and MR-ready performance evidence requirements. Uses an installed ncu-report-skill when a task needs detailed Nsight Compute collection and diagnosis.
Disciplined, reproducible loop for making an FLA kernel faster (Triton, Gluon, TileLang, CuTe) without ever breaking or gaming correctness. Synthesizes the task-contract / three-phase / iteration-protocol / silent-bug-catalog discipline of agent kernel-optimization frameworks (KDA, the MLSys FlashInfer contest workflow, AKO4ALL/AKO4X), and anchors all of it on FLA's frozen pytest (forward AND backward, under NaN poisoning) as the immutable correctness gate. Use when iterating on `fla/ops/**` performance over multiple rounds.