apple-app-intents
Skill9 starsChanged 2 months agoMITView on GitHub
Build and review iOS apps with App Intents, App Entities, App Shortcuts, and App Schemas for Siri and Apple Intelligence. Use when adding Siri actions, entity search, onscreen awareness, or migrating existing Shortcuts integrations. Includes SDK-aware implementation and validation; does not implement a replacement voice assistant.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- aka-kika/akakika-skillstheir own version, first seen Nov 1, 2026
- Dexter04056/apple-app-intents-skillword for word, first seen Sep 18, 2026
Other versions2 projects use a version of this.
Word for word · 1
- Dexter04056/apple-app-intents-skill★ 2100% same
Edited · 0
None found.
Wrote their own version · 1
- aka-kika/akakika-skills★ 11same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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
- Distinctive: line 19
Read platform-and-schemas.md before choosing a schema.
- Distinctive: line 41
Start with one meaningful action and its real data, then extend the map.
- Distinctive: line 45
Add the schema macros where they fit.
- Distinctive: line 48
Add onscreen entities, content transfer, and real UI-interaction donations when they improve the requested workflows.
- Distinctive: line 50
Use the bundled FieldNotes example as a pattern source, not a template to impose on unrelated apps.
- Distinctive: line 56
Read testing-and-release.md.
≈ “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.
Files
Apple App Intents
Build an app whose real capabilities are usable through Apple's system experiences. Treat intents as public interfaces to the app's existing services. Preserve the requested app, architecture, deployment target, and authorization boundaries.
Establish the target
Inspect the project, installed Xcode/SDK, minimum OS, existing intents, storage, navigation, and test targets. If useful, run python3 scripts/doctor.py /path/to/app from this skill's directory. Its output is an inventory, not a compliance verdict.
Read platform-and-schemas.md before choosing a schema. Distinguish:
- Custom App Intents and App Shortcuts for established system surfaces and invocation phrases.
- Schema-conforming entities and actions for Siri AI's supported domains.
- Features requiring newer SDKs, OS releases, enabled Apple Intelligence, languages, or regions.
The research baseline is 2026-09-10 / WWDC26. Recheck Apple's current symbol documentation and release notes when choosing APIs. Do not infer runtime availability from a keynote, macro conformance, or an older tutorial. If browsing is unavailable, label the affected claims unverified and use the installed SDK for compilation facts.
Design the integration
Produce a compact capability map before implementing substantial work:
| User action | Actual service operation | Entity/query | Exact schema or custom intent | Execution/authentication | Evidence needed |
| Open an item | Resolve ID and navigate | Persistent item lookup | Matching open schema or OpenIntent | Foreground | Cold launch opens that item |
Choose schemas by semantic fit and available system experience. Implement required groups completely. Do not relabel an unrelated object as a message, photo, or note just to obtain Siri discovery. If no supported schema matches, implement useful custom intents/App Shortcuts and explain the discovery limit.
For each selected schema, inspect its current generated template and symbol page. Record required property names, types, optionality, result types, and minimum OS. Do not infer names from patterns: the current Notes domain has create/update schemas, not a universal CRUD family. Extra optional parameters may serve Shortcuts but are not necessarily understood by Apple Intelligence.
Implement one working vertical slice
Start with one meaningful action and its real data, then extend the map. Use architecture.md for storage, dependencies, concurrency, authentication, and navigation.
- Model persisted records as
AppEntityvalues with stable identifiers and useful display representations. Queries rehydrate current authorized records by identifier; stale or inaccessible IDs never become fabricated records. Read entities-and-search.md. - Make
perform()call the same service as the UI, await the committed operation, and return truthful structured results. Validate inputs and permissions at execution time. Do not report success after only opening a screen or launching an unawaited task. - Add the schema macros where they fit. Compile early so metadata extraction and domain-completeness diagnostics can guide implementation. Preserve existing shortcut contracts during migration.
- Expose a small set of useful
AppShortcutentry points with app-name phrases and appropriate parameter summaries. Read shortcuts-and-dialog.md. - For discoverable persistent content, maintain a named Spotlight index through create, update, delete, access revocation, and repair. Type conformance alone does not index data.
- Add onscreen entities, content transfer, and real UI-interaction donations when they improve the requested workflows. Read context-and-transfer.md. Match the annotation API to the SDK and view structure.
Use the bundled FieldNotes example as a pattern source, not a template to impose on unrelated apps. It targets OS 27 and has explicitly documented scope and verification limits. Older deployment targets need their own availability strategy.
For large collections, cross-device identity, contextual relevance, longer work, or multiple execution processes, read advanced-2026.md before selecting newer APIs.
Verify the actual system contract
Read testing-and-release.md. Keep these evidence levels separate:
- Business-service behavior and persistence tests.
- Swift compilation and app-target metadata extraction, including schema diagnostics.
- AppIntentsTesting in an appropriately configured XCUITest bundle when available.
- Manual Shortcuts, Spotlight, and Siri tests on the intended device/OS.
Exercise missing parameters, ambiguous names, deleted IDs, denied authorization, offline failures, cold launches, repeated execution, index repair, and cancellation where relevant. Check that chained intents receive useful entities or values. Confirmations must happen before consequential side effects; refreshed ownership metadata complements application authorization rather than replacing it.
For bug reports, use troubleshooting.md to identify the failing layer before rewriting working code.
Deliver
Return the implemented capability map, modified files, exact build/test results, and remaining device or service checks. Label capabilities as implemented, compiled, system-tested, or device-verified only when the corresponding evidence exists. Provide concise example utterances and a Shortcuts flow the developer can reproduce.
Do not promise that every app or arbitrary custom action becomes available to unrestricted natural-language Siri requests. Do not add paid services, agent-specific APIs, telemetry, publishing, signing-account changes, or external submissions solely because this skill is active. This skill is free MIT-licensed guidance; it does not waive Apple's platform requirements.
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
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.

