refactor: replace custom qa runner with cali

.eas/workflows/agent-qa-mobile.yml

@@ -106,7 +106,10 @@ jobs:
       - uses: eas/install_node_modules
       - id: install_agent_device
         run: |
-          npm install -g agent-device@0.10.1
+          npm install -g agent-device@latest cali@0.4.0-2
+      - id: install_agent_device_skill
+        run: |
+          npx skills add callstackincubator/agent-device --agent codex --skill agent-device -y
       - uses: eas/download_build
         id: download_build
         with:
@@ -116,10 +119,6 @@ jobs:
         run: |
           bash ./scripts/agent-qa/provision-android-emulator.sh
       - id: run_agent_qa
-        env:
-          AGENT_DEVICE_SESSION: qa-android
-          AGENT_DEVICE_PLATFORM: android
-          AGENT_DEVICE_SESSION_LOCK: strip
         run: |
           bash ./scripts/agent-qa/run-and-export.sh "${{ steps.download_build.outputs.artifact_path }}"
 
@@ -150,20 +149,16 @@ jobs:
       - uses: eas/install_node_modules
       - id: install_agent_device
         run: |
-          npm install -g agent-device@0.10.1
+          npm install -g agent-device@latest cali@0.4.0-2
+      - id: install_agent_device_skill
+        run: |
+          npx skills add callstackincubator/agent-device --agent codex --skill agent-device -y
       - uses: eas/download_build
         id: download_build
         with:
           build_id: ${{ env.BUILD_ID }}
           extensions: [app]
-      - id: provision_ios_simulator
-        run: |
-          bash ./scripts/agent-qa/provision-ios-simulator.sh
       - id: run_agent_qa
-        env:
-          AGENT_DEVICE_SESSION: qa-ios
-          AGENT_DEVICE_PLATFORM: ios
-          AGENT_DEVICE_SESSION_LOCK: strip
         run: |
           bash ./scripts/agent-qa/run-and-export.sh "${{ steps.download_build.outputs.artifact_path }}"
 

README.md

@@ -1,12 +1,12 @@
 # EAS agent-device demo
 
-This repo is a minimal Expo + CNG example for running AI-assisted Android and iOS QA on EAS Workflows.
+This repo is a minimal Expo + CNG example for running AI-assisted Android and iOS QA on EAS Workflows with [`cali`](https://github.com/callstackincubator/cali).
 
 ## What it does
 
 - Reuses compatible Android and iOS simulator builds with `fingerprint` + `get-build` + `repack`
 - Falls back to a fresh `build` when the fingerprint changes
-- Runs a small Node.js QA agent built with the AI SDK `ToolLoopAgent`
+- Uses `cali qa` as the mobile QA agent runtime
 - Uses `agent-device` to drive the Android app and iOS simulator, take screenshots, and summarize findings
 - Posts one combined mobile QA summary back to the GitHub pull request with `github-comment`
 - Optionally uploads screenshots to Vercel Blob so the PR comment can link them
@@ -17,7 +17,8 @@ This repo is a minimal Expo + CNG example for running AI-assisted Android and iO
 
 - [eas.json](./eas.json)
 - [.eas/workflows/agent-qa-mobile.yml](./.eas/workflows/agent-qa-mobile.yml)
-- [scripts/agent-qa/index.ts](./scripts/agent-qa/index.ts)
+- [cali.config.json](./cali.config.json)
+- [scripts/agent-qa/run-and-export.sh](./scripts/agent-qa/run-and-export.sh)
 
 ## Required setup
 
@@ -34,38 +35,39 @@ Optional environment variables for the QA job:
 - `QA_MODEL`: Override the default model (`openai/gpt-5.4-mini`)
 - `BLOB_READ_WRITE_TOKEN`: Upload screenshots to Vercel Blob and include public links in the PR comment
 
+The repo-level [cali.config.json](./cali.config.json) points Cali at the bundled [`agent-device`](https://www.npmjs.com/package/agent-device) skills under `./node_modules/agent-device/skills`, so CI does not need a separate `~/.agents/skills` setup.
+
 ## Local smoke test
 
 ```bash
 npm install
-npx tsc --noEmit
+npx cali qa --help
 ```
 
-The workflow runner writes `section.md`, `status.txt`, and `report.json` to `artifacts/qa/` during execution. Temporary screenshots are written outside the workspace and uploaded to Vercel Blob when configured.
+The workflow runner writes `section.md`, `status.txt`, `report.json`, and `cali-context.json` to `artifacts/qa/` during execution. Screenshots are written to `artifacts/qa/screenshots` and uploaded to Vercel Blob when configured.
 
-To execute the runner directly with Node 24, provide the same environment variables the workflow sets:
+To execute the QA command directly, provide the same inputs that the workflow uses:
 
 Android:
 
 ```bash
 AI_GATEWAY_API_KEY=... \
-QA_PLATFORM=android \
-APP_PATH=/absolute/path/to/app.apk \
-APPLICATION_ID=dev.expo.easagentdevice \
-BUILD_ID=test-build \
-PR_JSON='{"number":1,"title":"Test PR","body":"Smoke test"}' \
-node ./scripts/agent-qa/index.ts
+./node_modules/.bin/cali qa \
+  --env local-android \
+  --artifact /absolute/path/to/app.apk \
+  --app-id dev.expo.easagentdevice \
+  --device ci-android \
+  --prompt "verify the updated welcome title"
 ```
 
 iOS simulator:
 
 ```bash
 AI_GATEWAY_API_KEY=... \
-QA_PLATFORM=ios \
-APP_PATH=/absolute/path/to/MyApp.app \
-APPLICATION_ID=dev.expo.easagentdevice \
-AGENT_DEVICE_IOS_DEVICE="iPhone 17" \
-BUILD_ID=test-build \
-PR_JSON='{"number":1,"title":"Test PR","body":"Smoke test"}' \
-node ./scripts/agent-qa/index.ts
+./node_modules/.bin/cali qa \
+  --env local-ios \
+  --artifact /absolute/path/to/MyApp.app \
+  --app-id dev.expo.easagentdevice \
+  --device "iPhone 17" \
+  --prompt "verify the updated welcome title"
 ```

cali.config.json

@@ -0,0 +1,13 @@
+{
+  "skillPaths": [
+    "./node_modules/agent-device/skills"
+  ],
+  "commands": {
+    "qa": {
+      "extraInstructions": [
+        "When you need to verify whether text is visible on screen, prefer `snapshot` over `snapshot -i`. Use `snapshot -i` mainly for interactive exploration and choosing refs.",
+        "When you save screenshots, use short descriptive file names and include matching screenshotLabels so downstream PR comments can label them clearly."
+      ]
+    }
+  }
+}