From 1c7ad118b7934271e8707c253c14edf82d656ad5 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 3 Aug 2025 06:19:00 +0000
Subject: [PATCH 1/7] Initial plan


From d2bf5937fef06f64dfd3aa14abb18ad6e0737676 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 3 Aug 2025 06:30:03 +0000
Subject: [PATCH 2/7] Add integration test suite with GitHub Actions workflow

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
---
 .github/workflows/integration.yml |  60 ++++++++
 .gitignore                        |   3 +
 DEV.md                            |  15 ++
 integration/README.md             |  76 ++++++++++
 integration/go.mod                |  11 ++
 integration/integration_test.go   | 221 ++++++++++++++++++++++++++++++
 6 files changed, 386 insertions(+)
 create mode 100644 .github/workflows/integration.yml
 create mode 100644 integration/README.md
 create mode 100644 integration/go.mod
 create mode 100644 integration/integration_test.go

diff --git a/.github/workflows/integration.yml b/.github/workflows/integration.yml
new file mode 100644
index 00000000..3bd344f0
--- /dev/null
+++ b/.github/workflows/integration.yml
@@ -0,0 +1,60 @@
+name: "Integration Tests"
+
+on:
+  push:
+    branches:
+      - 'main'
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  integration:
+    runs-on: ubuntu-latest
+    env:
+      GOPROXY: https://proxy.golang.org/,direct
+      GOPRIVATE: ""
+      GONOPROXY: ""
+      GONOSUMDB: github.com/github/*
+    steps:
+      - uses: actions/checkout@v4
+      
+      - uses: actions/setup-go@v5
+        with:
+          go-version: ">=1.22"
+          check-latest: true
+          
+      - name: Build gh-models binary
+        run: script/build
+        
+      - name: Run integration tests (without auth)
+        working-directory: integration
+        run: |
+          go mod tidy
+          go test -v -timeout=5m
+        env:
+          # Explicitly unset any GitHub tokens to test unauthenticated behavior
+          GITHUB_TOKEN: ""
+          GH_TOKEN: ""
+          
+      - name: Install gh CLI
+        if: github.event_name == 'workflow_dispatch'
+        run: |
+          curl -fsSL https://cli.github.com/packages/githubcli-archive-keyring.gpg | sudo dd of=/usr/share/keyrings/githubcli-archive-keyring.gpg
+          echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null
+          sudo apt update
+          sudo apt install gh
+          
+      - name: Run integration tests (with auth)
+        if: github.event_name == 'workflow_dispatch' 
+        working-directory: integration
+        run: |
+          gh auth login --with-token <<< "${{ secrets.GITHUB_TOKEN }}"
+          go test -v -timeout=10m
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
\ No newline at end of file
diff --git a/.gitignore b/.gitignore
index 54f9c6bc..396d8827 100644
--- a/.gitignore
+++ b/.gitignore
@@ -4,3 +4,6 @@
 /gh-models-linux-*
 /gh-models-windows-*
 /gh-models-android-*
+
+# Integration test dependencies
+integration/go.sum
diff --git a/DEV.md b/DEV.md
index 36c44fd1..f0cc739d 100644
--- a/DEV.md
+++ b/DEV.md
@@ -34,6 +34,21 @@ make vet  # to find suspicious constructs
 make tidy # to keep dependencies up-to-date
 ```
 
+### Integration Tests
+
+In addition to unit tests, we have integration tests that use the compiled binary to test against live endpoints:
+
+```shell
+# Build the binary first
+script/build
+
+# Run integration tests
+cd integration
+go test -v
+```
+
+Integration tests are located in the `integration/` directory and automatically skip tests requiring authentication when no GitHub token is available. See `integration/README.md` for more details.
+
 ## Releasing
 
 When upgrading or installing the extension using `gh extension upgrade github/gh-models` or
diff --git a/integration/README.md b/integration/README.md
new file mode 100644
index 00000000..5ebeb9a5
--- /dev/null
+++ b/integration/README.md
@@ -0,0 +1,76 @@
+# Integration Tests
+
+This directory contains integration tests for the `gh-models` CLI extension. These tests are separate from the unit tests and use the compiled binary to test actual functionality.
+
+## Overview
+
+The integration tests:
+- Use the compiled `gh-models` binary (not mocked clients)
+- Test basic functionality of each command (`list`, `run`, `view`, `eval`)
+- Are designed to work with or without GitHub authentication
+- Skip tests requiring live endpoints when authentication is unavailable
+- Keep assertions minimal to avoid brittleness
+
+## Running the Tests
+
+### Prerequisites
+
+1. Build the `gh-models` binary:
+   ```bash
+   cd ..
+   script/build
+   ```
+
+2. (Optional) Authenticate with GitHub CLI for full testing:
+   ```bash
+   gh auth login
+   ```
+
+### Running Locally
+
+From the integration directory:
+```bash
+go test -v
+```
+
+Without authentication, some tests will be skipped:
+```
+=== RUN   TestIntegrationHelp
+--- PASS: TestIntegrationHelp (0.05s)
+=== RUN   TestIntegrationList
+    integration_test.go:90: Skipping integration test - no GitHub authentication available
+--- SKIP: TestIntegrationList (0.04s)
+```
+
+With authentication, all tests should run and test live endpoints.
+
+## CI/CD
+
+The integration tests run automatically on pushes to `main` via the GitHub Actions workflow `.github/workflows/integration.yml`.
+
+The workflow:
+1. Builds the binary
+2. Runs tests without authentication (tests basic functionality)
+3. On manual dispatch, can also run with authentication for full testing
+
+## Test Structure
+
+Each test follows this pattern:
+- Check for binary existence (skip if not built)
+- Check for authentication (skip live endpoint tests if unavailable)
+- Execute the binary with specific arguments
+- Verify basic output format and success/failure
+
+Tests are intentionally simple and focus on:
+- Commands execute without errors
+- Help text is present and correctly formatted
+- Basic output format is as expected
+- Authentication requirements are respected
+
+## Adding New Tests
+
+When adding new commands or features:
+1. Add a corresponding integration test
+2. Follow the existing pattern of checking authentication
+3. Keep assertions minimal but meaningful
+4. Ensure tests work both with and without authentication
\ No newline at end of file
diff --git a/integration/go.mod b/integration/go.mod
new file mode 100644
index 00000000..3e104b8f
--- /dev/null
+++ b/integration/go.mod
@@ -0,0 +1,11 @@
+module github.com/github/gh-models/integration
+
+go 1.22
+
+require github.com/stretchr/testify v1.10.0
+
+require (
+	github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc // indirect
+	github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
+)
diff --git a/integration/integration_test.go b/integration/integration_test.go
new file mode 100644
index 00000000..3de61e07
--- /dev/null
+++ b/integration/integration_test.go
@@ -0,0 +1,221 @@
+package integration
+
+import (
+	"os"
+	"os/exec"
+	"path/filepath"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+)
+
+const (
+	binaryName      = "gh-models"
+	timeoutDuration = 30 * time.Second
+)
+
+// getBinaryPath returns the path to the compiled gh-models binary
+func getBinaryPath(t *testing.T) string {
+	wd, err := os.Getwd()
+	require.NoError(t, err)
+
+	// Binary should be in the parent directory
+	binaryPath := filepath.Join(filepath.Dir(wd), binaryName)
+
+	// Check if binary exists
+	if _, err := os.Stat(binaryPath); os.IsNotExist(err) {
+		t.Skipf("Binary %s not found. Run 'script/build' first.", binaryPath)
+	}
+
+	return binaryPath
+}
+
+// hasAuthToken checks if GitHub authentication is available
+func hasAuthToken() bool {
+	// Check if gh CLI is available and authenticated
+	cmd := exec.Command("gh", "auth", "status")
+	return cmd.Run() == nil
+}
+
+// runCommand executes the gh-models binary with given arguments
+func runCommand(t *testing.T, args ...string) (stdout, stderr string, err error) {
+	binaryPath := getBinaryPath(t)
+
+	cmd := exec.Command(binaryPath, args...)
+	cmd.Env = os.Environ()
+
+	// Set timeout
+	done := make(chan error, 1)
+	var stdoutBytes, stderrBytes []byte
+
+	go func() {
+		stdoutBytes, err = cmd.Output()
+		if err != nil {
+			if exitError, ok := err.(*exec.ExitError); ok {
+				stderrBytes = exitError.Stderr
+			}
+		}
+		done <- err
+	}()
+
+	select {
+	case err = <-done:
+		return string(stdoutBytes), string(stderrBytes), err
+	case <-time.After(timeoutDuration):
+		if cmd.Process != nil {
+			cmd.Process.Kill()
+		}
+		t.Fatalf("Command timed out after %v", timeoutDuration)
+		return "", "", nil
+	}
+}
+
+func TestIntegrationHelp(t *testing.T) {
+	stdout, stderr, err := runCommand(t, "--help")
+
+	// Help should always work, even without auth
+	require.NoError(t, err, "stderr: %s", stderr)
+	require.Contains(t, stdout, "GitHub Models CLI extension")
+	require.Contains(t, stdout, "Available Commands:")
+	require.Contains(t, stdout, "list")
+	require.Contains(t, stdout, "run")
+	require.Contains(t, stdout, "view")
+	require.Contains(t, stdout, "eval")
+}
+
+func TestIntegrationList(t *testing.T) {
+	if !hasAuthToken() {
+		t.Skip("Skipping integration test - no GitHub authentication available")
+	}
+
+	stdout, stderr, err := runCommand(t, "list")
+
+	if err != nil {
+		t.Logf("List command failed. stdout: %s, stderr: %s", stdout, stderr)
+		// If the command fails due to auth issues, skip the test
+		if strings.Contains(stderr, "authentication") || strings.Contains(stderr, "token") {
+			t.Skip("Skipping - authentication issue")
+		}
+		require.NoError(t, err, "List command should succeed with valid auth")
+	}
+
+	// Basic verification that list command produces expected output format
+	require.NotEmpty(t, stdout, "List should produce output")
+	// Should contain some indication of models or table headers
+	lowerOut := strings.ToLower(stdout)
+	hasExpectedContent := strings.Contains(lowerOut, "model") ||
+		strings.Contains(lowerOut, "name") ||
+		strings.Contains(lowerOut, "id") ||
+		strings.Contains(lowerOut, "display")
+	require.True(t, hasExpectedContent, "List output should contain model information")
+}
+
+func TestIntegrationListHelp(t *testing.T) {
+	stdout, stderr, err := runCommand(t, "list", "--help")
+
+	require.NoError(t, err, "stderr: %s", stderr)
+	require.Contains(t, stdout, "Returns a list of models")
+	require.Contains(t, stdout, "Usage:")
+}
+
+func TestIntegrationView(t *testing.T) {
+	if !hasAuthToken() {
+		t.Skip("Skipping integration test - no GitHub authentication available")
+	}
+
+	// First get a model to view
+	listOut, _, listErr := runCommand(t, "list")
+	if listErr != nil {
+		t.Skip("Cannot run view test - list command failed")
+	}
+
+	// Extract a model name from list output (this is basic parsing)
+	lines := strings.Split(listOut, "\n")
+	var modelName string
+	for _, line := range lines {
+		line = strings.TrimSpace(line)
+		// Look for lines that might contain model IDs (containing forward slash)
+		if strings.Contains(line, "/") && !strings.HasPrefix(line, "Usage:") &&
+			!strings.HasPrefix(line, "gh models") && line != "" {
+			// Try to extract what looks like a model ID
+			fields := strings.Fields(line)
+			for _, field := range fields {
+				if strings.Contains(field, "/") {
+					modelName = field
+					break
+				}
+			}
+			if modelName != "" {
+				break
+			}
+		}
+	}
+
+	if modelName == "" {
+		t.Skip("Could not extract model name from list output")
+	}
+
+	stdout, stderr, err := runCommand(t, "view", modelName)
+
+	if err != nil {
+		t.Logf("View command failed for model %s. stdout: %s, stderr: %s", modelName, stdout, stderr)
+		// If the command fails due to auth issues, skip the test
+		if strings.Contains(stderr, "authentication") || strings.Contains(stderr, "token") {
+			t.Skip("Skipping - authentication issue")
+		}
+		require.NoError(t, err, "View command should succeed with valid model")
+	}
+
+	// Basic verification that view command produces expected output
+	require.NotEmpty(t, stdout, "View should produce output")
+	lowerOut := strings.ToLower(stdout)
+	hasExpectedContent := strings.Contains(lowerOut, "model") ||
+		strings.Contains(lowerOut, "name") ||
+		strings.Contains(lowerOut, "description") ||
+		strings.Contains(lowerOut, "publisher")
+	require.True(t, hasExpectedContent, "View output should contain model details")
+}
+
+func TestIntegrationViewHelp(t *testing.T) {
+	stdout, stderr, err := runCommand(t, "view", "--help")
+
+	require.NoError(t, err, "stderr: %s", stderr)
+	require.Contains(t, stdout, "Returns details about the specified model")
+	require.Contains(t, stdout, "Usage:")
+}
+
+func TestIntegrationRunHelp(t *testing.T) {
+	stdout, stderr, err := runCommand(t, "run", "--help")
+
+	require.NoError(t, err, "stderr: %s", stderr)
+	require.Contains(t, stdout, "Prompts the specified model")
+	require.Contains(t, stdout, "Usage:")
+}
+
+func TestIntegrationEvalHelp(t *testing.T) {
+	stdout, stderr, err := runCommand(t, "eval", "--help")
+
+	require.NoError(t, err, "stderr: %s", stderr)
+	require.Contains(t, stdout, "Runs evaluation tests against a model")
+	require.Contains(t, stdout, "Usage:")
+}
+
+// TestIntegrationRun tests the run command with a simple prompt
+// This test is more limited since it requires actual model inference
+func TestIntegrationRun(t *testing.T) {
+	if !hasAuthToken() {
+		t.Skip("Skipping integration test - no GitHub authentication available")
+	}
+
+	// We'll test with a very simple prompt to minimize cost and time
+	// Using a basic model and short prompt
+	stdout, _, err := runCommand(t, "run", "--help")
+	require.NoError(t, err, "Run help should work")
+
+	// For now, just verify the help works.
+	// A full test would require setting up a prompt and model,
+	// which might be expensive for CI
+	require.Contains(t, stdout, "Prompts the specified model")
+}

From 78b652d9bf43ffc4a9697d0cc6a31f08bfbb5e28 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 3 Aug 2025 08:23:49 +0000
Subject: [PATCH 3/7] Address feedback: add build target, update workflow, and
 simplify

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
---
 .github/workflows/integration.yml | 22 +++-------------------
 DEV.md                            |  2 +-
 Makefile                          |  5 +++++
 3 files changed, 9 insertions(+), 20 deletions(-)

diff --git a/.github/workflows/integration.yml b/.github/workflows/integration.yml
index 3bd344f0..75c7b75d 100644
--- a/.github/workflows/integration.yml
+++ b/.github/workflows/integration.yml
@@ -8,6 +8,7 @@ on:
 
 permissions:
   contents: read
+  models: read
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
@@ -30,7 +31,7 @@ jobs:
           check-latest: true
           
       - name: Build gh-models binary
-        run: script/build
+        run: make build
         
       - name: Run integration tests (without auth)
         working-directory: integration
@@ -40,21 +41,4 @@ jobs:
         env:
           # Explicitly unset any GitHub tokens to test unauthenticated behavior
           GITHUB_TOKEN: ""
-          GH_TOKEN: ""
-          
-      - name: Install gh CLI
-        if: github.event_name == 'workflow_dispatch'
-        run: |
-          curl -fsSL https://cli.github.com/packages/githubcli-archive-keyring.gpg | sudo dd of=/usr/share/keyrings/githubcli-archive-keyring.gpg
-          echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null
-          sudo apt update
-          sudo apt install gh
-          
-      - name: Run integration tests (with auth)
-        if: github.event_name == 'workflow_dispatch' 
-        working-directory: integration
-        run: |
-          gh auth login --with-token <<< "${{ secrets.GITHUB_TOKEN }}"
-          go test -v -timeout=10m
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
\ No newline at end of file
+          GH_TOKEN: ""
\ No newline at end of file
diff --git a/DEV.md b/DEV.md
index f0cc739d..fb33d9bc 100644
--- a/DEV.md
+++ b/DEV.md
@@ -40,7 +40,7 @@ In addition to unit tests, we have integration tests that use the compiled binar
 
 ```shell
 # Build the binary first
-script/build
+make build
 
 # Run integration tests
 cd integration
diff --git a/Makefile b/Makefile
index 898120db..c8d50d01 100644
--- a/Makefile
+++ b/Makefile
@@ -1,6 +1,11 @@
 check: fmt vet tidy test
 .PHONY: check
 
+build:
+	@echo "==> building gh-models binary <=="
+	script/build
+.PHONY: build
+
 fmt:
 	@echo "==> running Go format <=="
 	gofmt -s -l -w .

From 296a540f52d2a608654da01a8f11a1a9c2afd36c Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 3 Aug 2025 12:52:51 +0000
Subject: [PATCH 4/7] Add org tests, pass GitHub token in workflow, and add
 integration Makefile target

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
---
 .github/workflows/integration.yml |  6 ++----
 Makefile                          |  5 +++++
 integration/integration_test.go   | 26 ++++++++++++++++++++++++++
 3 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/integration.yml b/.github/workflows/integration.yml
index 75c7b75d..7d9477f4 100644
--- a/.github/workflows/integration.yml
+++ b/.github/workflows/integration.yml
@@ -33,12 +33,10 @@ jobs:
       - name: Build gh-models binary
         run: make build
         
-      - name: Run integration tests (without auth)
+      - name: Run integration tests
         working-directory: integration
         run: |
           go mod tidy
           go test -v -timeout=5m
         env:
-          # Explicitly unset any GitHub tokens to test unauthenticated behavior
-          GITHUB_TOKEN: ""
-          GH_TOKEN: ""
\ No newline at end of file
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
\ No newline at end of file
diff --git a/Makefile b/Makefile
index c8d50d01..7194239c 100644
--- a/Makefile
+++ b/Makefile
@@ -6,6 +6,11 @@ build:
 	script/build
 .PHONY: build
 
+integration: build
+	@echo "==> running integration tests <=="
+	cd integration && go test -v -timeout=5m
+.PHONY: integration
+
 fmt:
 	@echo "==> running Go format <=="
 	gofmt -s -l -w .
diff --git a/integration/integration_test.go b/integration/integration_test.go
index 3de61e07..0e9704a2 100644
--- a/integration/integration_test.go
+++ b/integration/integration_test.go
@@ -219,3 +219,29 @@ func TestIntegrationRun(t *testing.T) {
 	// which might be expensive for CI
 	require.Contains(t, stdout, "Prompts the specified model")
 }
+
+// TestIntegrationRunWithOrg tests the run command with --org flag
+func TestIntegrationRunWithOrg(t *testing.T) {
+	if !hasAuthToken() {
+		t.Skip("Skipping integration test - no GitHub authentication available")
+	}
+
+	// Test run command with --org flag (using help to avoid expensive API calls)
+	stdout, _, err := runCommand(t, "run", "--org", "test-org", "--help")
+	require.NoError(t, err, "Run help with --org should work")
+	require.Contains(t, stdout, "Prompts the specified model")
+	require.Contains(t, stdout, "--org string")
+}
+
+// TestIntegrationEvalWithOrg tests the eval command with --org flag
+func TestIntegrationEvalWithOrg(t *testing.T) {
+	if !hasAuthToken() {
+		t.Skip("Skipping integration test - no GitHub authentication available")
+	}
+
+	// Test eval command with --org flag (using help to avoid expensive API calls)
+	stdout, _, err := runCommand(t, "eval", "--org", "test-org", "--help")
+	require.NoError(t, err, "Eval help with --org should work")
+	require.Contains(t, stdout, "Runs evaluation tests against a model")
+	require.Contains(t, stdout, "--org string")
+}

From 175460997f13fb2b40eb23e40f7e2755e2606207 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 3 Aug 2025 14:49:15 +0000
Subject: [PATCH 5/7] Change integration test binary name to gh-models-test to
 avoid confusion

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
---
 .gitignore                      | 1 +
 Makefile                        | 2 ++
 integration/integration_test.go | 2 +-
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/.gitignore b/.gitignore
index 396d8827..8f1fad44 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,5 +1,6 @@
 /gh-models
 /gh-models.exe
+/gh-models-test
 /gh-models-darwin-*
 /gh-models-linux-*
 /gh-models-windows-*
diff --git a/Makefile b/Makefile
index 7194239c..ab83ff62 100644
--- a/Makefile
+++ b/Makefile
@@ -7,6 +7,8 @@ build:
 .PHONY: build
 
 integration: build
+	@echo "==> building gh-models-test binary for integration tests <=="
+	go build -o gh-models-test main.go
 	@echo "==> running integration tests <=="
 	cd integration && go test -v -timeout=5m
 .PHONY: integration
diff --git a/integration/integration_test.go b/integration/integration_test.go
index 0e9704a2..5fe67103 100644
--- a/integration/integration_test.go
+++ b/integration/integration_test.go
@@ -12,7 +12,7 @@ import (
 )
 
 const (
-	binaryName      = "gh-models"
+	binaryName      = "gh-models-test"
 	timeoutDuration = 30 * time.Second
 )
 

From 5614b048ffbf92ad1341b256eb18f61a3d0cd2e9 Mon Sep 17 00:00:00 2001
From: Peli de Halleux <pelikhan@users.noreply.github.com>
Date: Sun, 3 Aug 2025 21:12:44 +0000
Subject: [PATCH 6/7] Refactor integration tests: remove unused auth token
 check and streamline test cases

---
 Makefile                        |   4 +-
 integration/integration_test.go | 168 +++-----------------------------
 2 files changed, 12 insertions(+), 160 deletions(-)

diff --git a/Makefile b/Makefile
index ab83ff62..12bde1ff 100644
--- a/Makefile
+++ b/Makefile
@@ -7,10 +7,8 @@ build:
 .PHONY: build
 
 integration: build
-	@echo "==> building gh-models-test binary for integration tests <=="
-	go build -o gh-models-test main.go
 	@echo "==> running integration tests <=="
-	cd integration && go test -v -timeout=5m
+	cd integration && go mod tidy && go test -v -timeout=5m
 .PHONY: integration
 
 fmt:
diff --git a/integration/integration_test.go b/integration/integration_test.go
index 5fe67103..0b7bfb28 100644
--- a/integration/integration_test.go
+++ b/integration/integration_test.go
@@ -32,13 +32,6 @@ func getBinaryPath(t *testing.T) string {
 	return binaryPath
 }
 
-// hasAuthToken checks if GitHub authentication is available
-func hasAuthToken() bool {
-	// Check if gh CLI is available and authenticated
-	cmd := exec.Command("gh", "auth", "status")
-	return cmd.Run() == nil
-}
-
 // runCommand executes the gh-models binary with given arguments
 func runCommand(t *testing.T, args ...string) (stdout, stderr string, err error) {
 	binaryPath := getBinaryPath(t)
@@ -72,26 +65,8 @@ func runCommand(t *testing.T, args ...string) (stdout, stderr string, err error)
 	}
 }
 
-func TestIntegrationHelp(t *testing.T) {
-	stdout, stderr, err := runCommand(t, "--help")
-
-	// Help should always work, even without auth
-	require.NoError(t, err, "stderr: %s", stderr)
-	require.Contains(t, stdout, "GitHub Models CLI extension")
-	require.Contains(t, stdout, "Available Commands:")
-	require.Contains(t, stdout, "list")
-	require.Contains(t, stdout, "run")
-	require.Contains(t, stdout, "view")
-	require.Contains(t, stdout, "eval")
-}
-
-func TestIntegrationList(t *testing.T) {
-	if !hasAuthToken() {
-		t.Skip("Skipping integration test - no GitHub authentication available")
-	}
-
+func TestList(t *testing.T) {
 	stdout, stderr, err := runCommand(t, "list")
-
 	if err != nil {
 		t.Logf("List command failed. stdout: %s, stderr: %s", stdout, stderr)
 		// If the command fails due to auth issues, skip the test
@@ -105,143 +80,22 @@ func TestIntegrationList(t *testing.T) {
 	require.NotEmpty(t, stdout, "List should produce output")
 	// Should contain some indication of models or table headers
 	lowerOut := strings.ToLower(stdout)
-	hasExpectedContent := strings.Contains(lowerOut, "model") ||
-		strings.Contains(lowerOut, "name") ||
-		strings.Contains(lowerOut, "id") ||
-		strings.Contains(lowerOut, "display")
+	hasExpectedContent := strings.Contains(lowerOut, "openai/gpt-4.1")
 	require.True(t, hasExpectedContent, "List output should contain model information")
 }
 
-func TestIntegrationListHelp(t *testing.T) {
-	stdout, stderr, err := runCommand(t, "list", "--help")
-
-	require.NoError(t, err, "stderr: %s", stderr)
-	require.Contains(t, stdout, "Returns a list of models")
-	require.Contains(t, stdout, "Usage:")
-}
-
-func TestIntegrationView(t *testing.T) {
-	if !hasAuthToken() {
-		t.Skip("Skipping integration test - no GitHub authentication available")
-	}
-
-	// First get a model to view
-	listOut, _, listErr := runCommand(t, "list")
-	if listErr != nil {
-		t.Skip("Cannot run view test - list command failed")
-	}
-
-	// Extract a model name from list output (this is basic parsing)
-	lines := strings.Split(listOut, "\n")
-	var modelName string
-	for _, line := range lines {
-		line = strings.TrimSpace(line)
-		// Look for lines that might contain model IDs (containing forward slash)
-		if strings.Contains(line, "/") && !strings.HasPrefix(line, "Usage:") &&
-			!strings.HasPrefix(line, "gh models") && line != "" {
-			// Try to extract what looks like a model ID
-			fields := strings.Fields(line)
-			for _, field := range fields {
-				if strings.Contains(field, "/") {
-					modelName = field
-					break
-				}
-			}
-			if modelName != "" {
-				break
-			}
-		}
-	}
-
-	if modelName == "" {
-		t.Skip("Could not extract model name from list output")
-	}
-
-	stdout, stderr, err := runCommand(t, "view", modelName)
-
-	if err != nil {
-		t.Logf("View command failed for model %s. stdout: %s, stderr: %s", modelName, stdout, stderr)
-		// If the command fails due to auth issues, skip the test
-		if strings.Contains(stderr, "authentication") || strings.Contains(stderr, "token") {
-			t.Skip("Skipping - authentication issue")
-		}
-		require.NoError(t, err, "View command should succeed with valid model")
-	}
-
-	// Basic verification that view command produces expected output
-	require.NotEmpty(t, stdout, "View should produce output")
-	lowerOut := strings.ToLower(stdout)
-	hasExpectedContent := strings.Contains(lowerOut, "model") ||
-		strings.Contains(lowerOut, "name") ||
-		strings.Contains(lowerOut, "description") ||
-		strings.Contains(lowerOut, "publisher")
-	require.True(t, hasExpectedContent, "View output should contain model details")
-}
-
-func TestIntegrationViewHelp(t *testing.T) {
-	stdout, stderr, err := runCommand(t, "view", "--help")
-
-	require.NoError(t, err, "stderr: %s", stderr)
-	require.Contains(t, stdout, "Returns details about the specified model")
-	require.Contains(t, stdout, "Usage:")
-}
-
-func TestIntegrationRunHelp(t *testing.T) {
-	stdout, stderr, err := runCommand(t, "run", "--help")
-
-	require.NoError(t, err, "stderr: %s", stderr)
-	require.Contains(t, stdout, "Prompts the specified model")
-	require.Contains(t, stdout, "Usage:")
-}
-
-func TestIntegrationEvalHelp(t *testing.T) {
-	stdout, stderr, err := runCommand(t, "eval", "--help")
-
-	require.NoError(t, err, "stderr: %s", stderr)
-	require.Contains(t, stdout, "Runs evaluation tests against a model")
-	require.Contains(t, stdout, "Usage:")
-}
-
-// TestIntegrationRun tests the run command with a simple prompt
+// TestRun tests the run command with a simple prompt
 // This test is more limited since it requires actual model inference
-func TestIntegrationRun(t *testing.T) {
-	if !hasAuthToken() {
-		t.Skip("Skipping integration test - no GitHub authentication available")
-	}
-
-	// We'll test with a very simple prompt to minimize cost and time
-	// Using a basic model and short prompt
-	stdout, _, err := runCommand(t, "run", "--help")
-	require.NoError(t, err, "Run help should work")
-
-	// For now, just verify the help works.
-	// A full test would require setting up a prompt and model,
-	// which might be expensive for CI
-	require.Contains(t, stdout, "Prompts the specified model")
+func TestRun(t *testing.T) {
+	stdout, _, err := runCommand(t, "run", "openai/gpt-4.1-nano", "say 'pain' in french")
+	require.NoError(t, err, "Run should work")
+	require.Contains(t, strings.ToLower(stdout), "pain")
 }
 
 // TestIntegrationRunWithOrg tests the run command with --org flag
-func TestIntegrationRunWithOrg(t *testing.T) {
-	if !hasAuthToken() {
-		t.Skip("Skipping integration test - no GitHub authentication available")
-	}
-
+func TestRunWithOrg(t *testing.T) {
 	// Test run command with --org flag (using help to avoid expensive API calls)
-	stdout, _, err := runCommand(t, "run", "--org", "test-org", "--help")
-	require.NoError(t, err, "Run help with --org should work")
-	require.Contains(t, stdout, "Prompts the specified model")
-	require.Contains(t, stdout, "--org string")
-}
-
-// TestIntegrationEvalWithOrg tests the eval command with --org flag
-func TestIntegrationEvalWithOrg(t *testing.T) {
-	if !hasAuthToken() {
-		t.Skip("Skipping integration test - no GitHub authentication available")
-	}
-
-	// Test eval command with --org flag (using help to avoid expensive API calls)
-	stdout, _, err := runCommand(t, "eval", "--org", "test-org", "--help")
-	require.NoError(t, err, "Eval help with --org should work")
-	require.Contains(t, stdout, "Runs evaluation tests against a model")
-	require.Contains(t, stdout, "--org string")
+	stdout, _, err := runCommand(t, "run", "openai/gpt-4.1-nano", "say 'pain' in french", "--org", "github")
+	require.NoError(t, err, "Run should work")
+	require.Contains(t, strings.ToLower(stdout), "pain")
 }

From 92529c593c681cf2e1b21aa998132cfc7801c0f6 Mon Sep 17 00:00:00 2001
From: Peli de Halleux <pelikhan@users.noreply.github.com>
Date: Sun, 3 Aug 2025 21:14:31 +0000
Subject: [PATCH 7/7] Refactor integration workflow: streamline Go setup and
 remove unnecessary build steps

---
 .github/workflows/integration.yml | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/.github/workflows/integration.yml b/.github/workflows/integration.yml
index 7d9477f4..44376043 100644
--- a/.github/workflows/integration.yml
+++ b/.github/workflows/integration.yml
@@ -24,19 +24,11 @@ jobs:
       GONOSUMDB: github.com/github/*
     steps:
       - uses: actions/checkout@v4
-      
-      - uses: actions/setup-go@v5
+      - name: Setup Go
+        uses: actions/setup-go@v5
         with:
-          go-version: ">=1.22"
-          check-latest: true
-          
-      - name: Build gh-models binary
-        run: make build
-        
+          go-version-file: 'go.mod'
       - name: Run integration tests
-        working-directory: integration
-        run: |
-          go mod tidy
-          go test -v -timeout=5m
+        run: make integration
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
\ No newline at end of file