agentleFS
SajidisrarSajidisrar/ios-claude-skills/skills/ios-app-intents-siri/SKILL.md

ios-app-intents-siri

Skill1 starsChanged 17 days agoMITView on GitHub

Expose an iOS app's content and actions to Siri, Spotlight, Shortcuts, widgets and Apple Intelligence through the App Intents framework — app entities, intents, assistant/semantic schemas, on-screen content (View Annotations), and testing with AppIntentsTesting. Use when adding Siri or Shortcuts support, making app content findable in Spotlight or usable by Apple Intelligence, migrating from SiriKit intents, or when an intent or entity stops compiling after an SDK update.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    SiriKit: If the app still uses SiriKit intents, plan the move to App Intents.

    line 40

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

App Intents: Siri, Spotlight and Apple Intelligence

App Intents is the one integration point the system uses to find your app's content and perform its actions — Siri, Spotlight, Shortcuts, widgets, controls and Apple Intelligence all read the same intents and entities. Model them once, well.

API names below are from Apple's WWDC26 iOS guide and the iOS 27 release notes. Check the App Intents documentation for exact signatures and availability before writing code; this area changes every release.

1. Decide what to expose

List the app's nouns (entities: an order, a note, a contact) and verbs (intents: create, open, mark done). Expose what people would ask for by voice or search, not every screen. For each, note: the identifier that stays stable, the display representation, and which data must never leave the app (sensitive fields stay out of entities).

2. Model entities and intents

  • AppEntity for each noun, with a stable id, a display representation, and an entity query so the system can resolve references ("my last order").
  • AppIntent for each verb: parameters, a perform() that calls the same repository/use-case layer the UI uses (never duplicate business logic in an intent), and dialog/result values.
  • Schemas. Where Apple defines a schema for your domain (assistant schemas; in iOS 27, entity schemas and intent schemas), conform to it — it lets Siri and Apple Intelligence understand your content without you defining phrases, and lets entities be added to the Spotlight semantic index with attribution. Schemas gain properties between releases: in iOS 27 an entity conforming to the photos asset schema stopped compiling until the new properties were adopted behind an availability check. Re-check schema conformances on every SDK upgrade (ios-sdk-upgrade).
  • On-screen content. iOS 27 adds a View Annotations API that maps views to entities, so people can act on what's on screen conversationally. Annotate the screens where "this one" is a natural request (a detail view), not every list row.
  • SiriKit. If the app still uses SiriKit intents, plan the move to App Intents; new Siri and Apple Intelligence features are built on App Intents.

3. Shortcuts, Spotlight and widgets

  • Provide App Shortcuts for the two or three most common actions so they work with no setup.
  • Index entities with Core Spotlight only for content the user would search for; update or delete index entries when data changes or the user signs out.
  • Reuse intents for widget configuration and interactive widget/control buttons (ios-widgetkit-live-activities) rather than writing parallel logic.

4. Privacy and authentication

  • An intent can run while the device is locked or the app isn't in the foreground. Decide per intent whether it needs authentication or an unlocked device, and enforce it.
  • Entities you donate or index are visible to system features. Never include data the user would be surprised to see in Spotlight or a Siri answer. Update ios-privacy-compliance answers if new data is surfaced.

5. Test

  • Unit-test perform() through the same test doubles as the rest of the app (ios-unit-tests).
  • iOS 27 adds an AppIntentsTesting framework for validating integrations through system pathways without UI automation — use it for entity resolution and intent execution where available.
  • Try each intent by voice, in Shortcuts and in Spotlight on a device, signed in and signed out, and with the device locked.

Verification

  • Exposed entities and intents match what users would actually ask for; sensitive data excluded.
  • Every intent calls the app's existing use-case layer; no duplicated business logic.
  • Schema conformances used where Apple defines one, and re-checked against the current SDK.
  • Each intent's authentication/locked-device behaviour is decided and enforced.
  • Spotlight index entries are updated and removed when data changes or on sign-out.
  • Intents are unit-tested, and checked by voice, in Shortcuts and in Spotlight on a device.
  • Exact API names and availability were checked against Apple's current documentation.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.