Add AppKit support and databricks-agent-skills integration (#356)

* Apps: Add AppKit, and prefer it over other app frameworks

* rename local databricks-apps skill to databricks-apps-python

* install: enable renaming skills from remote repos

* Drop agent skills for jobs and pipelines

Remove databricks-jobs:databricks-jobs-bundles and databricks-pipelines
from agent skills in both install.sh and install.ps1.

Co-authored-by: Isaac

* Rename databricks-app-python → databricks-apps-python everywhere

Completes the pre-existing-bug fix called out in the PR description. The
skill directory `databricks-apps-python` (plural) is the canonical name;
this updates all remaining stale singular references (.test eval skill,
routing ground truth, scorer, baselines, run_app_eval docs, builder-app
system prompt + UI, MCP tool docstring, install_skills.sh, related-skill
links in lakebase-provisioned/bundles SKILL.md).

Verified .test/tests/test_scorers.py: 36/36 pass.

Co-authored-by: Isaac

* fix: ACE multi-model review findings on agent-skill installer

Two issues caught in parallel review by GPT 5.4 and Gemini 3.1 Pro:

1. **Upstream skill name mismatch** (HIGH, GPT 5.4): The agent skill
   at `databricks/databricks-agent-skills` is named `databricks-core`,
   not `databricks`. Previously `--skills-profile app-developer` would
   silently fail to install one of three advertised skills.
   Fix: use the source:install-name syntax this PR already introduces
   — `databricks-core:databricks` keeps the local install name while
   pointing at the correct upstream directory.

2. **Partial-download false success** (CRITICAL, Gemini 3.1 Pro):
   `ok_flag` was initialized to 0 and flipped to 1 on the first
   successful file download, never reset on subsequent failures.
   Any single successful file (even an examples/*.py) marked the
   skill as installed, leaving a broken skill on disk. Same pattern
   in install.ps1 with $okFlag.
   Fix: initialize the flag to 1 (true) and only flip to 0 (false)
   on any single download failure — a skill installs cleanly iff
   every file lands.

Not addressed (acceptable today, worth a follow-up):
- Unauthenticated GitHub API rate limit (60/hr) — fine for single-
  user installs, can bite CI.
- `grep '\.'` heuristic to filter trees vs blobs — brittle but works
  for current skills (all files have extensions).
- Sequential vs parallel curl — perf nit, not a bug.

Co-authored-by: Isaac

* fix: address ACE multi-model review findings

GPT 5.4 xhigh and Gemini 3.1 Pro both flagged the new short agent-skill
names triggering latent bugs in the install scripts. Three fixes:

1. install.sh `_is_preselected`: `grep -qw` treats `-` as a word boundary,
   so checking for `databricks` would falsely match `databricks-jobs`,
   `databricks-apps`, etc. Strip `source:` prefix from each preselected
   entry and use `grep -Fxq` for exact whole-line equality. Same fix
   applied to the deselection cleanup at the cleanup loop.

2. install.ps1: `$preselected -contains "databricks"` is exact equality on
   array elements, so it never matched the `"databricks-core:databricks"`
   entry seeded by the app-developer profile — the Agent: Databricks
   checkbox was never auto-preselected. Normalize `$preselected` by
   stripping `source:` prefixes once before the menu is built.

3. install.sh/ps1: the "Agent skills (N) -> ..." success line ran
   unconditionally, even when the GitHub tree fetch failed or every
   per-skill download warned and was rolled back. Track an
   `agent_success`/`$agentSuccess` counter and only print the success
   line when all selected agent skills installed; print a warning when
   only some succeeded.

Co-authored-by: Isaac

* fix: agent-skills installer regex matches live GitHub API + clears stale dirs

Two more ACE review findings, both real:

1. The path-extraction regex `"path":"skills/..."` (no space) does NOT match
   the GitHub tree API's actual response, which is pretty-printed as
   `"path": "skills/..."` (space after colon). Result: the installer would
   warn "Could not fetch agent skill" for every entry and install nothing.
   Fix: collapse the JSON whitespace before regex extraction, and switch
   from the `grep '\.'` heuristic to matching the adjacent
   `"type": "blob"` field so directory entries are correctly skipped.

2. Reinstalls reused the existing $dest_dir / $destDir without clearing
   it, so files removed upstream would persist locally across upgrades.
   Fix: `rm -rf` / `Remove-Item -Recurse` the destination before each
   skill is downloaded.

Co-authored-by: Isaac

* fix: resolve_skills bucketing + 0-of-N agent install summary

ACE iteration 2 findings:

- Gemini caught that the `grep -qw` word-boundary bug I fixed in
  `_is_preselected` still bit the resolve_skills() bucketing chain
  (lines 918/920/922). Passing `--skills databricks` (a valid agent
  install-name) would match `databricks-app-apx` via the hyphen
  word-boundary on the APX branch and get misclassified. Switched
  all three buckets to exact match via `tr ' ' '\n' | grep -Fxq`,
  and rewrote the source:install-name lookup on line 923 with `-E`
  and a clean anchored alternation.

- GPT caught that when the GitHub tree fetch fails entirely
  (`agent_tree` empty / `$agentTree` null), the summary branch
  emits nothing — neither OK nor "only N of M installed" — even
  though zero of N agent skills were actually installed. Added an
  `else` branch that emits a `0 of $agent_count installed` warning.

Out of scope (flagged but not addressed):
- GPT raised legacy-name compatibility for `databricks-app-python`
  → `databricks-apps-python`. The rename is intentional per the
  product decision; back-compat aliasing is a separate concern.

Co-authored-by: Isaac

Author: jamesbroadhead

…ines/databricks-app-python/baseline.yaml …nes/databricks-apps-python/baseline.yaml.test/baselines/databricks-app-python/baseline.yaml renamed to .test/baselines/databricks-apps-python/baseline.yaml .test/baselines/databricks-app-python/baseline.yaml renamed to .test/baselines/databricks-apps-python/baseline.yaml

@@ -10,6 +10,6 @@ results_summary:
   status: RUNNING
   success: true
 run_id: 1fb85925959b453090dce7d3317559cf
-skill_name: databricks-app-python
+skill_name: databricks-apps-python
 test_count: 1
 timestamp: '2026-02-12T20:17:59Z'

.test/scripts/run_app_eval.py

@@ -9,8 +9,8 @@
 5. Cleans up all test apps
 
 Usage:
-    DATABRICKS_CONFIG_PROFILE=ffe python run_app_eval.py databricks-app-python [--test-ids id1 id2]
-    DATABRICKS_CONFIG_PROFILE=ffe python run_app_eval.py databricks-app-python --keep  # Don't delete apps after
+    DATABRICKS_CONFIG_PROFILE=ffe python run_app_eval.py databricks-apps-python [--test-ids id1 id2]
+    DATABRICKS_CONFIG_PROFILE=ffe python run_app_eval.py databricks-apps-python --keep  # Don't delete apps after
 """
 
 import argparse
@@ -226,7 +226,7 @@ def detect_framework_yaml(python_code: str) -> str:
 
 def main():
     parser = argparse.ArgumentParser(description="Run app skill integration tests on Databricks")
-    parser.add_argument("skill_name", help="Name of skill to evaluate (e.g., databricks-app-python)")
+    parser.add_argument("skill_name", help="Name of skill to evaluate (e.g., databricks-apps-python)")
     parser.add_argument("--test-ids", nargs="+", help="Specific test IDs to run")
     parser.add_argument("--keep", action="store_true", help="Don't delete apps after testing")
     args = parser.parse_args()

.test/skills/_routing/ground_truth.yaml

@@ -148,7 +148,7 @@ test_cases:
     inputs:
       prompt: "Create a Streamlit app that shows sales data"
     expectations:
-      expected_skills: ["databricks-app-python"]
+      expected_skills: ["databricks-apps-python"]
       is_multi_skill: false
     metadata:
       category: "single_skill"
@@ -159,7 +159,7 @@ test_cases:
     inputs:
       prompt: "Build a Dash app with interactive charts"
     expectations:
-      expected_skills: ["databricks-app-python"]
+      expected_skills: ["databricks-apps-python"]
       is_multi_skill: false
     metadata:
       category: "single_skill"
@@ -170,7 +170,7 @@ test_cases:
     inputs:
       prompt: "Create a Gradio app for testing my ML model"
     expectations:
-      expected_skills: ["databricks-app-python"]
+      expected_skills: ["databricks-apps-python"]
       is_multi_skill: false
     metadata:
       category: "single_skill"
@@ -181,7 +181,7 @@ test_cases:
     inputs:
       prompt: "Build a FastAPI app that serves data from a warehouse"
     expectations:
-      expected_skills: ["databricks-app-python"]
+      expected_skills: ["databricks-apps-python"]
       is_multi_skill: false
     metadata:
       category: "single_skill"
@@ -192,7 +192,7 @@ test_cases:
     inputs:
       prompt: "Create a Reflex app for managing inventory"
     expectations:
-      expected_skills: ["databricks-app-python"]
+      expected_skills: ["databricks-apps-python"]
       is_multi_skill: false
     metadata:
       category: "single_skill"
@@ -205,7 +205,7 @@ test_cases:
       prompt: "Build a Dash app and deploy it using DABs"
     expectations:
       expected_skills:
-        - "databricks-app-python"
+        - "databricks-apps-python"
         - "databricks-bundles"
       is_multi_skill: true
     metadata:
@@ -218,7 +218,7 @@ test_cases:
       prompt: "Create a Streamlit app that stores data in Lakebase"
     expectations:
       expected_skills:
-        - "databricks-app-python"
+        - "databricks-apps-python"
         - "databricks-lakebase-provisioned"
       is_multi_skill: true
     metadata:
@@ -231,7 +231,7 @@ test_cases:
       prompt: "Build a Gradio app that queries a model serving endpoint"
     expectations:
       expected_skills:
-        - "databricks-app-python"
+        - "databricks-apps-python"
         - "model-serving"
       is_multi_skill: true
     metadata:
@@ -246,7 +246,7 @@ test_cases:
     expectations:
       expected_skills:
         - "databricks-app-apx"
-        - "databricks-app-python"
+        - "databricks-apps-python"
       is_multi_skill: true
     metadata:
       category: "multi_skill"

…ls/databricks-app-python/candidates.yaml …s/databricks-apps-python/candidates.yaml.test/skills/databricks-app-python/candidates.yaml renamed to .test/skills/databricks-apps-python/candidates.yaml .test/skills/databricks-app-python/candidates.yaml renamed to .test/skills/databricks-apps-python/candidates.yaml

@@ -1,6 +1,6 @@
 skill:
-  name: "databricks-app-python"
-  source_path: "databricks-skills/databricks-app-python"
+  name: "databricks-apps-python"
+  source_path: "databricks-skills/databricks-apps-python"
   description: "Python Databricks Apps: Dash, Streamlit, Gradio, Flask, FastAPI, Reflex"
 
 tool_modules: [apps, serving]

…/databricks-app-python/ground_truth.yaml …databricks-apps-python/ground_truth.yaml.test/skills/databricks-app-python/ground_truth.yaml renamed to .test/skills/databricks-apps-python/ground_truth.yaml .test/skills/databricks-app-python/ground_truth.yaml renamed to .test/skills/databricks-apps-python/ground_truth.yaml

@@ -32,7 +32,7 @@
         "fastapi react",
         "react frontend",
     ],
-    "databricks-app-python": [
+    "databricks-apps-python": [
         "python app",
         "streamlit",
         "dash",

…ills/databricks-app-python/manifest.yaml …lls/databricks-apps-python/manifest.yaml.test/skills/databricks-app-python/manifest.yaml renamed to .test/skills/databricks-apps-python/manifest.yaml .test/skills/databricks-app-python/manifest.yaml renamed to .test/skills/databricks-apps-python/manifest.yaml

@@ -70,34 +70,34 @@ def test_detect_genie(self):
         assert "databricks-agent-bricks" in skills
 
     def test_detect_app_python_streamlit(self):
-        """Test detection of databricks-app-python via Streamlit."""
+        """Test detection of databricks-apps-python via Streamlit."""
         prompt = "Create a Streamlit app that shows sales data"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_app_python_dash(self):
-        """Test detection of databricks-app-python via Dash."""
+        """Test detection of databricks-apps-python via Dash."""
         prompt = "Build a Dash app with interactive charts"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_app_python_gradio(self):
-        """Test detection of databricks-app-python via Gradio."""
+        """Test detection of databricks-apps-python via Gradio."""
         prompt = "Create a Gradio app for testing my ML model"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_app_python_fastapi(self):
-        """Test detection of databricks-app-python via FastAPI."""
+        """Test detection of databricks-apps-python via FastAPI."""
         prompt = "Build a FastAPI app that serves data from a warehouse"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_app_python_reflex(self):
-        """Test detection of databricks-app-python via Reflex."""
+        """Test detection of databricks-apps-python via Reflex."""
         prompt = "Create a Reflex app for managing inventory"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_app_apx(self):
         """Test detection of databricks-app-apx."""
@@ -109,13 +109,13 @@ def test_detect_fastapi_react_matches_both(self):
         """Test that 'FastAPI React' matches both APX and Python app skills.
 
         'fastapi react' triggers APX, while bare 'fastapi' also triggers
-        databricks-app-python. This is intentional — the router sees both
+        databricks-apps-python. This is intentional — the router sees both
         and picks the best fit.
         """
         prompt = "Create a FastAPI React app for my dashboard"
         skills = detect_skills_from_prompt(prompt)
         assert "databricks-app-apx" in skills
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
 
     def test_detect_lakebase(self):
         """Test detection of databricks-lakebase-provisioned skill."""
@@ -140,14 +140,14 @@ def test_detect_multi_app_lakebase(self):
         """Test detection of app + lakebase."""
         prompt = "Create a Streamlit app that stores data in Lakebase"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
         assert "databricks-lakebase-provisioned" in skills
 
     def test_detect_multi_app_serving(self):
         """Test detection of app + model serving."""
         prompt = "Build a Gradio app that queries a model serving endpoint"
         skills = detect_skills_from_prompt(prompt)
-        assert "databricks-app-python" in skills
+        assert "databricks-apps-python" in skills
         assert "databricks-model-serving" in skills
 
     def test_detect_no_match(self):
@@ -171,7 +171,7 @@ def test_all_skills_have_triggers(self):
         expected_skills = [
             "databricks-spark-declarative-pipelines",
             "databricks-app-apx",
-            "databricks-app-python",
+            "databricks-apps-python",
             "databricks-bundles",
             "databricks-python-sdk",
             "databricks-jobs",

.test/src/skill_test/scorers/routing.py

@@ -179,7 +179,7 @@ Skills provide specialized guidance for Databricks development tasks. They are m
 Skills include:
 - **databricks-bundles**: DABs configuration
 - **databricks-app-apx**: Full-stack apps with APX framework (FastAPI + React)
-- **databricks-app-python**: Python apps with Dash, Streamlit, Flask
+- **databricks-apps-python**: Python apps with Dash, Streamlit, Flask
 - **databricks-python-sdk**: Python SDK patterns
 - **databricks-mlflow-evaluation**: MLflow evaluation and trace analysis
 - **databricks-spark-declarative-pipelines**: Spark Declarative Pipelines (SDP) development
@@ -313,7 +313,7 @@ Skills are loaded from `../databricks-skills/` and filtered by the `ENABLED_SKIL
 - `databricks-spark-declarative-pipelines`: SDP/DLT pipeline development
 - `databricks-synthetic-data-gen`: Creating test datasets
 - `databricks-app-apx`: Full-stack apps with React (APX framework)
-- `databricks-app-python`: Python apps with Dash, Streamlit, Flask
+- `databricks-apps-python`: Python apps with Dash, Streamlit, Flask
 
 **Adding custom skills:**
 1. Create a new directory in `../databricks-skills/`

.test/tests/test_scorers.py

@@ -30,7 +30,7 @@ env:
   # =============================================================================
   # Comma-separated list of skills to enable
   - name: ENABLED_SKILLS
-    value: "databricks-asset-bundles,databricks-agent-bricks,databricks-aibi-dashboards,databricks-app-apx,databricks-app-python,databricks-config,databricks-docs,databricks-jobs,databricks-python-sdk,databricks-unity-catalog,mlflow-evaluation,spark-declarative-pipelines,synthetic-data-generation,unstructured-pdf-generation"
+    value: "databricks-asset-bundles,databricks-agent-bricks,databricks-aibi-dashboards,databricks-app-apx,databricks-apps-python,databricks-config,databricks-docs,databricks-jobs,databricks-python-sdk,databricks-unity-catalog,mlflow-evaluation,spark-declarative-pipelines,synthetic-data-generation,unstructured-pdf-generation"
   - name: SKILLS_ONLY_MODE
     value: "false"