Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Build and maintain mobile applications end-to-end with OpenClaw, including requirement shaping, architecture, implementation, debugging, testing, and release...
Build and maintain mobile applications end-to-end with OpenClaw, including requirement shaping, architecture, implementation, debugging, testing, and release...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
Prioritize shipping working, testable increments over large unverified rewrites. Keep solutions cross-platform by default (iOS + Android) unless the user requests platform-specific behavior. Reuse existing project patterns before introducing new abstractions. Prefer readable, strongly typed code and small, composable modules. Validate changes with commands the user can run locally.
Capture the user goal as a concrete deliverable. Confirm constraints: stack, timeline, supported platforms, auth, backend, offline needs, and notifications. Convert ambiguous requests into explicit acceptance criteria.
Detect whether this is: a greenfield app, a feature addition, a bugfix, a refactor/performance pass, release hardening. Tailor the implementation depth to the detected mode.
Identify impacted screens, state, navigation, data layer, and native capabilities. Define the smallest safe implementation slice. List verification commands before coding.
Create or update one coherent unit at a time (UI, hook/viewmodel, API client, schema, tests). Keep business logic out of view layers where possible. Avoid introducing unused dependencies.
Run relevant checks (typecheck, lint, tests, build/start commands). Report what passed, what failed, and what was not run. Summarize changed files and key behavior updates.
Default stack: React Native + Expo + TypeScript. State strategy: use existing app pattern first (context/store/query library). Networking: typed API client boundaries and defensive parsing. Forms: explicit validation and clear user error states. Navigation: preserve current routing conventions. Styling: follow existing design system/tokens; avoid one-off inline styles.
Confirm loading, empty, success, and error states exist. Confirm touch targets are usable and layout adapts to small screens. Confirm text wraps correctly and avoids clipped content. Confirm accessibility labels/roles on interactive elements. Confirm no crashes from undefined/null edge cases. Confirm async actions have visible progress and failure handling.
Minimize unnecessary re-renders in lists and heavy screens. Memoize expensive derived values when profiling shows need. Keep bundle impact low; remove dead imports. Defer non-critical work from initial screen render.
Treat API contracts as versioned interfaces. Avoid breaking existing clients without a migration path. Add backward-compatible fields/functions when behavior changes. Keep serialization/deserialization logic centralized.
Verify app config, permissions, deep links, and environment variables. Ensure icons, splash assets, and bundle identifiers/package names are consistent. Confirm release build commands and signing prerequisites are documented. Provide a short release checklist for iOS and Android when requested.
Reproduce first with clear steps. Isolate whether issue is UI, state, network, storage, or native bridge. Add narrow instrumentation/logging only where needed. Fix root cause, then remove temporary debugging noise. Add regression coverage for high-risk bugs.
For each completed task, provide: What changed. Why the approach was chosen. Exact files touched. Validation commands run and outcomes. Remaining risks or follow-ups.
Do not fabricate API responses, device behavior, or test results. Do not claim a task is complete without verification evidence. Do not perform destructive data/schema changes without explicit user approval. Do not break existing navigation or auth flows to satisfy local changes.
Scaffold app shell. Implement core navigation and one vertical feature slice. Add typed API boundary and mock/real data switch. Add baseline test and lint/typecheck pass.
Reuse existing screen/module pattern. Add minimal data contract changes. Add/adjust tests for new behavior.
Write reproduction notes. Fix smallest root cause. Add regression test if feasible.
Code helpers, APIs, CLIs, browser automation, testing, and developer operations.
Largest current source with strong distribution and engagement signals.