More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    EAS Update: Over-the-Air App Updates

    1

    Ship Expo JS and asset fixes to installed apps without a new binary: the delivery contract, the update-safe decision, staging-first publishing, and the debug ladder.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    EAS Update: Over-the-Air App Updates

    EAS Update: Over-the-Air App Updates

    Example session with this skill installed

    My Expo app has a login bug: users with a saved session from the previous version get a blank screen. The fix is a small JS change. The store review would take days — can I push this over the air, and how do I prove it actually reached users?

    • Read your context and instructions
    • Compiled the eas update

    OTA PUBLISH RUN — login blank-screen fix, worked through the skill.

    1. SHIPPABLE? The fix touches only JS session-restoration code: no native library, no config plugin, no runtime-version change. Compatibility table verdict: update-safe. Had it needed a new native module, the answer would be a new build, full stop.

    2. CONTRACT CHECK: Installed builds run runtime 1.4.0 on channel production; the production channel points at the production branch. The publish targets exactly that runtime and platform set — anything else would silently reach nobody.

    3. STAGING FIRST: Published to staging with the message "Fix blank screen for restored sessions from 1.3" and the staging environment, after checking the installed CLI's update --help for current flags. On the staging release build: fully terminate, reopen (embedded update shows, download happens in background), terminate, reopen — fix visible on the second cold launch. That's expected arrival behavior, not a failure.

    4. PRODUCTION: Previous production group id recorded from update:list as the rollback target, explicit approval obtained, same change published to production with the same message. Publish record filled in: channel, branch, runtime 1.4.0, group ids, staging check noted.

    5. PROOF: Next check of update health shows adoption climbing on runtime 1.4.0 and blank-screen session reports stopped. If the fix had misbehaved, the recorded previous group would be republished — the rollback path was written down before it was needed.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Your fix is ready, but a store release takes days and your users are crashing now. EAS Update can deliver the fix over the air — if the change is actually update-safe, if the project's channels, branches, and runtime versions line up, and if you validate the way updates really arrive on devices instead of the way you wish they did. This skill teaches the delivery contract first: build, update, branch, channel, and runtime version, and the one rule that decides whether an installed app ever sees your publish. It sets the project up through the CLI rather than hand-written configuration, then gives you the compatibility table that answers "can this ship as an update?" before every publish — pure JS and assets yes, native changes and incidental runtime-policy edits never. The publish flow is staging-first with a real message, a two-cold-launch validation on a release build, a written publish record (including the rollback target group id), and explicit human approval before production — never inferred from a branch name. A complete worked example walks a checkout crash fix from staging to production with the health check afterward. When an update "didn't appear," the six-rung debugging ladder works the causes in order — published where you think, build match, channel knowledge, channel-to-branch mapping, the two cold launches, native logs — instead of random republishing. Edge cases cover mixed runtime fleets, channel remap accidents, and environment mismatches, and the rollback section makes sure the recovery path is written down before you need it.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Mobile Development.

    • Updates cannot deliver native changes: new native libraries, native config changes, and SDK upgrades still require a new binary and store submission.
    • The skill covers EAS Update specifically; it does not set up EAS Build profiles or store submission, and it does not apply to non-Expo React Native update systems.
    • Arrival timing on user devices depends on their launch behavior; the skill proves delivery to a build, not to every individual user instantly.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 24 days ago

    Software
    expoNode.jseas-cli

    Permissions

    Terminal / Shell
    Listed24 days ago
    Updated21 days ago

    What's inside

    Frequently Asked Questions

    Probably not. Release builds launch the cached update while downloading the new one in the background, so the fix typically appears on the second full cold launch — terminate the app completely, don't just background it. If it's still missing after two cold launches, the debugging ladder in the skill finds the real cause, usually a runtime-version or channel mismatch.

    No. Size doesn't matter — any native change is outside what an update can deliver, and forcing it risks shipping a bundle the installed native runtime can't execute. The compatibility table in the skill draws this line before every publish.

    Yes — the staging pass is short, and it's the only place the two-cold-launch arrival behavior gets validated on a release build before users are involved. The skill allows fast-tracking the staging soak for urgent fixes; it never allows skipping it.

    New in Mobile Development