feat(shared): lint rule flags aiSearch indexes missing columns

[MarioCadenas]

What

Adds an appkit lint rule, ai-search-index-requires-columns, that flags an aiSearch plugin config whose indexes have no usable columns.

Why

In production the plugin does not auto-discover an index's columns — that's dev-only (ai-search.ts _autoDiscoverColumns, gated on NODE_ENV === "development"). So a config that omits columns works in dev but fails at query time in prod (Field 'columns' must be specified). appkit lint is the pre-deploy static catch, complementing the existing dev nudges (auto-discovery banner + manifest scaffolding.rules).

How

aiSearch({ indexes: { a: {...}, b: {...} } }) — "every index has non-empty columns" can't be expressed as a single ast-grep pattern, so this adds an optional find?: (root) => SgNode[] escape hatch to the Rule interface. Flat pattern rules and this structural rule share the same reporting path. lintFile now delegates to an exported lintSource() so rules can be tested against in-memory source.

Flagged (fail in prod):

• bare aiSearch(), aiSearch({}), aiSearch({ indexes: {} }) — all fall back to the env-seeded default index, which has no columns
• any index object missing columns, or with columns: []

Passed (can't statically prove bad — no false positives):

• columns: SOME_CONST, indexes: buildIndexes(), aiSearch(dynamicConfig), spread index objects ({ ...base })

The violation points at the offending index alias, not the whole call.

Test plan

• packages/shared/src/cli/commands/lint.test.ts — 14 vitest cases covering every pass/flag branch above
• pnpm exec vitest run lint — 14/14 pass
• pnpm -r typecheck — clean
• pnpm exec biome check on changed files — clean
• Manual: appkit lint over a temp server.ts with a bad aiSearch() flags it and exits 1

Manual testing + output

No columns added to the plugin:

createApp({
  plugins: [
    aiSearch(),
    server(),
  ],
}).catch(console.error);

or this

createApp({
  plugins: [
 aiSearch({ indexes: { default: { columns: [] } } }),
    server(),
  ],
}).catch(console.error);

will produce this output

➜ npm run lint:ast-grep

> appkit-pr-513-app@1.0.0 lint:ast-grep
> appkit lint

Scanning 11 TypeScript files...

Found 1 violation(s):

server/server.ts:6:5
  ai-search-index-requires-columns: AI Search index has no `columns`; it will fail in production (columns are auto-discovered only in dev). Set `columns` explicitly on each index.
  > aiSearch()

Columns added

createApp({
  plugins: [
    aiSearch({ indexes: { default: { columns: ['my_column', 'column'] } } }),
    server(),
  ],
}).catch(console.error);

➜ npm run lint:ast-grep

> appkit-pr-513-app@1.0.0 lint:ast-grep
> appkit lint

Scanning 11 TypeScript files...

No ast-grep lint violations found.

[github-actions]

📦 Bundle size report

Compared against bundle-size-baseline.json (main).

✅ No size changes vs the baseline.

@databricks/appkit

npm tarball (packed): 827 KB — gzipped download (dist + bin; excludes release-only docs/NOTICE).

dist	raw	gzip
JS (runtime)	855 KB	298 KB
Type declarations	314 KB	108 KB
Source maps	1.7 MB	557 KB
Other	11 KB	3.7 KB
Total	2.8 MB	967 KB

Per-entry composition (own code — deps external (as shipped))

Entry	Initial (gz)	Lazy (gz)	Total (gz)	node_modules (min)	Own code (min)
.	87 KB	2.5 KB	90 KB	external	285 KB
./beta	48 KB	459 B	48 KB	external	140 KB
./type-generator	20 KB	0 B	20 KB	external	57 KB

Chunks:

Entry	Chunk	Load	Size (gz)
.	index.js	initial	83 KB
.	utils.js	initial	4.0 KB
.	remote-tunnel-manager.js	lazy	2.5 KB
./beta	beta.js	initial	32 KB
./beta	stream-manager.js	initial	5.8 KB
./beta	wide-event-emitter.js	initial	3.2 KB
./beta	databricks.js	initial	3.0 KB
./beta	configuration.js	initial	2.1 KB
./beta	service-context.js	initial	1.3 KB
./beta	client.js	initial	431 B
./beta	client-options.js	initial	220 B
./beta	supervisor-api.js	lazy	194 B
./beta	databricks.js	lazy	142 B
./beta	index.js	lazy	123 B
./type-generator	index.js	initial	20 KB

@databricks/appkit-ui

npm tarball (packed): 309 KB — gzipped download (dist + bin; excludes release-only docs/NOTICE).

dist	raw	gzip
JS (runtime)	362 KB	121 KB
Type declarations	209 KB	76 KB
Source maps	693 KB	227 KB
CSS	16 KB	3.3 KB
Total	1.3 MB	427 KB

Per-entry composition (consumer bundle — deps bundled, peerDeps external)

Entry	Initial (gz)	Lazy (gz)	Total (gz)	node_modules (min)	Own code (min)
./js	4.3 KB	49 KB	54 KB	208 KB	12 KB
./js/beta	20 B	0 B	20 B	0 B	0 B
./react	429 KB	49 KB	478 KB	1.3 MB	168 KB
./react/beta	1.0 KB	0 B	1.0 KB	0 B	1.9 KB

Chunks:

Entry	Chunk	Load	Size (gz)
./js	index.js	initial	4.2 KB
./js	chunk	initial	120 B
./js	apache-arrow	lazy	49 KB
./js/beta	beta.js	initial	20 B
./react	index.js	initial	427 KB
./react	tslib	initial	2.1 KB
./react	apache-arrow	lazy	49 KB
./react/beta	beta.js	initial	1.0 KB

[github-actions]

🤖 AppKit PR bot

🔬 Run evals

Start an eval for this PR from the evals-monitor app: Go to Evals Monitor →

📦 Try this PR's app template

Scaffolds a new app from this PR's SDK build. Run it in any folder (requires the GitHub CLI — gh auth login — and the Databricks CLI):

gh run download 31169096353 -R databricks/appkit -n appkit-template-0.55.1-pr.1948e4a-feat-ai-search-columns-lint-513 -D appkit-pr-513 \
  && unzip -o "appkit-pr-513/appkit-template-0.55.1-pr.1948e4a-feat-ai-search-columns-lint-513.zip" -d "appkit-pr-513" \
  && databricks apps init --template "appkit-pr-513"

The template pins @databricks/appkit and @databricks/appkit-ui to tarballs built from this branch, so the scaffolded app runs against this PR's code.

[atilafassina]

i think i found one edge-case with a spread operator, but let's not defer this plugin on that. It delivers well despite a small false-positive that arguably would be against best practice anyway 🐑 it