diff --git a/.claude/skills/refresh-databricks-skills/SKILL.md b/.claude/skills/refresh-databricks-skills/SKILL.md index 47395dd..b78561e 100644 --- a/.claude/skills/refresh-databricks-skills/SKILL.md +++ b/.claude/skills/refresh-databricks-skills/SKILL.md @@ -1,6 +1,6 @@ --- name: refresh-databricks-skills -description: Use when Databricks skills need updating, user asks to refresh or sync skills from upstream, or skills seem outdated compared to the ai-dev-kit repo +description: Use when Databricks skills need updating, user asks to refresh or sync skills from upstream, or skills seem outdated compared to the databricks-agent-skills repo --- # Refresh Databricks Skills @@ -9,7 +9,9 @@ description: Use when Databricks skills need updating, user asks to refresh or s Pulls the latest Databricks skills from the upstream source repo and replaces all existing Databricks skills in the project while preserving non-Databricks skills (e.g., superpowers workflow skills). -**Source repo:** `https://github.com/databricks-solutions/ai-dev-kit` (path: `databricks-skills/`) +**Source repo:** `https://github.com/databricks/databricks-agent-skills` (paths: `skills/` for stable + `experimental/` for experimental). + +The upstream content was previously [`databricks-solutions/ai-dev-kit`](https://github.com/databricks-solutions/ai-dev-kit) (now deprecated). DAS is the canonical source going forward — stable skills live under `skills/` and experimental skills (the verbatim a-d-k snapshot) live under `experimental/`. ## When to Use @@ -21,28 +23,33 @@ Pulls the latest Databricks skills from the upstream source repo and replaces al 1. **Clone the upstream repo** (shallow clone for speed): ```bash - git clone --depth 1 https://github.com/databricks-solutions/ai-dev-kit.git $TMPDIR/ai-dev-kit + git clone --depth 1 https://github.com/databricks/databricks-agent-skills.git $TMPDIR/das ``` -2. **Identify non-Databricks skills to preserve.** These are the superpowers workflow skills that live alongside Databricks skills. List them by checking which directories in `.claude/skills/` do NOT have a matching folder in the upstream `databricks-skills/` directory. Common superpowers skills include: `brainstorming`, `dispatching-parallel-agents`, `executing-plans`, `finishing-a-development-branch`, `receiving-code-review`, `requesting-code-review`, `subagent-driven-development`, `systematic-debugging`, `test-driven-development`, `using-git-worktrees`, `using-superpowers`, `verification-before-completion`, `writing-plans`, `writing-skills`. Also preserve any other project-specific skills (like this one: `refresh-databricks-skills`). +2. **Identify non-Databricks skills to preserve.** These are the superpowers workflow skills that live alongside Databricks skills. List them by checking which directories in `.claude/skills/` do NOT have a matching folder under `skills/` or `experimental/` in the upstream repo. Common superpowers skills include: `brainstorming`, `dispatching-parallel-agents`, `executing-plans`, `finishing-a-development-branch`, `receiving-code-review`, `requesting-code-review`, `subagent-driven-development`, `systematic-debugging`, `test-driven-development`, `using-git-worktrees`, `using-superpowers`, `verification-before-completion`, `writing-plans`, `writing-skills`. Also preserve any other project-specific skills (like this one: `refresh-databricks-skills`). 3. **Remove old Databricks skills** from `.claude/skills/`, keeping all non-Databricks skills identified above. -4. **Copy new Databricks skills** from the cloned repo. Copy every directory under `databricks-skills/` except `TEMPLATE`: +4. **Copy new Databricks skills** from the cloned repo. Copy every directory under both `skills/` (stable) and `experimental/`: ```bash SKILLS_DIR=".claude/skills" - UPSTREAM="$TMPDIR/ai-dev-kit/databricks-skills" - for dir in "$UPSTREAM"/databricks-* "$UPSTREAM"/spark-*; do - [ -d "$dir" ] && cp -r "$dir" "$SKILLS_DIR/$(basename "$dir")" + UPSTREAM_STABLE="$TMPDIR/das/skills" + UPSTREAM_EXPERIMENTAL="$TMPDIR/das/experimental" + for src in "$UPSTREAM_STABLE" "$UPSTREAM_EXPERIMENTAL"; do + for dir in "$src"/databricks-* "$src"/spark-*; do + [ -d "$dir" ] && cp -r "$dir" "$SKILLS_DIR/$(basename "$dir")" + done done ``` + Note: name collisions across `skills/` and `experimental/` are impossible — DAS's `scripts/skills.py` rejects them at generate-time. + 5. **Clean up** the cloned repo: ```bash - rm -rf $TMPDIR/ai-dev-kit + rm -rf $TMPDIR/das ``` -6. **Report** the count of skills added, removed, and updated. +6. **Report** the count of skills added, removed, and updated, and call out whether each added skill came from `skills/` (stable) or `experimental/`. ## After Refreshing @@ -54,6 +61,8 @@ databricks apps deploy --source-code-path --profile ## Common Mistakes +- **Pointing at the deprecated upstream:** Old versions of this skill cloned `databricks-solutions/ai-dev-kit`. That repo is deprecated and no longer receives skill updates — always use `databricks/databricks-agent-skills`. +- **Forgetting `experimental/`:** DAS splits skills into `skills/` (stable, reviewed) and `experimental/` (best-effort a-d-k import). Both directories must be copied or you'll regress functionality. - **Deleting non-Databricks skills:** Always identify and preserve superpowers and project-specific skills before removing anything. - **Forgetting this skill itself:** `refresh-databricks-skills` must be preserved during the refresh. - **Not using `--depth 1`:** Full clone is slow and unnecessary. Always shallow clone. diff --git a/CLAUDE.md b/CLAUDE.md index 5ccac7f..9e634d1 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -88,7 +88,7 @@ Real-time terminal I/O over **WebSocket** (Flask-SocketIO) with automatic **HTTP ## Credits -- Databricks skills from [databricks-solutions/ai-dev-kit](https://github.com/databricks-solutions/ai-dev-kit) +- Databricks skills from [databricks/databricks-agent-skills](https://github.com/databricks/databricks-agent-skills) (canonical source; the legacy [databricks-solutions/ai-dev-kit](https://github.com/databricks-solutions/ai-dev-kit) is deprecated) - Development workflow skills from [obra/superpowers](https://github.com/obra/superpowers) # things to remember diff --git a/README.md b/README.md index 139da30..191cf72 100644 --- a/README.md +++ b/README.md @@ -175,7 +175,7 @@ This template repo opens that vision up for every Databricks user — no IDE set
🧠 All 39 Skills -### Databricks Skills (25) — [ai-dev-kit](https://github.com/databricks-solutions/ai-dev-kit) +### Databricks Skills (25) — [databricks-agent-skills](https://github.com/databricks/databricks-agent-skills) | Category | Skills | |----------|--------|