Refactor PromptPex model handling by changing pointer fields to values; update related tests and documentation for consistency

Author: pelikhan

.github/copilot-instructions.md

@@ -8,7 +8,7 @@ This repository implements the GitHub Models CLI extension (`gh models`), enabli
 ### Building and Testing
 
 - `make build`: Compiles the CLI binary
-- `make check`: Runs format, vet, tidy, and tests. Always run when you are done with changes.
+- `make check`: Runs format, vet, tidy, tests, golang-ci. Always run when you are done with changes. Use this command to validate that the build and the tests are still ok.
 - `make test`: Runs the tests.
 
 ### Command Structure
@@ -33,7 +33,7 @@ This repository implements the GitHub Models CLI extension (`gh models`), enabli
 - **Local build**: `make build` or `script/build` (creates `gh-models` binary)
 - **Cross-platform**: `script/build all|windows|linux|darwin` for release builds
 - **Testing**: `make check` runs format, vet, tidy, and tests. Use `go test ./...` directly for faster iteration
-- **Quality gates**: `make fmt vet tidy test` - required before commits
+- **Quality gates**: `make check` - required before commits
 
 ### Authentication & Setup
 - Extension requires `gh auth login` before use - unauthenticated clients show helpful error messages

cmd/generate/generate.go

@@ -121,7 +121,7 @@ func ParseFlags(cmd *cobra.Command, options *PromptPexOptions) error {
 	}
 
 	if groundtruthModel, _ := flags.GetString("groundtruth-model"); groundtruthModel != "" {
-		options.Models.Groundtruth = &groundtruthModel
+		options.Models.Groundtruth = groundtruthModel
 	}
 
 	return nil

cmd/generate/generate_test.go

@@ -79,8 +79,7 @@ func TestParseFlags(t *testing.T) {
 			name: "groundtruth model flag",
 			args: []string{"--groundtruth-model", "openai/gpt-4o"},
 			validate: func(t *testing.T, opts *PromptPexOptions) {
-				require.NotNil(t, opts.Models.Groundtruth)
-				require.Equal(t, "openai/gpt-4o", *opts.Models.Groundtruth)
+				require.Equal(t, "openai/gpt-4o", opts.Models.Groundtruth)
 			},
 		},
 	}

cmd/generate/options.go

@@ -10,9 +10,9 @@ func GetDefaultOptions() *PromptPexOptions {
 		MaxRulesPerTestGen: util.Ptr(3),
 		Verbose:            util.Ptr(false),
 		Models: &PromptPexModelAliases{
-			Rules:       util.Ptr("openai/gpt-4o"),
-			Tests:       util.Ptr("openai/gpt-4o"),
-			Groundtruth: util.Ptr("openai/gpt-4o"),
+			Rules:       "openai/gpt-4o",
+			Tests:       "openai/gpt-4o",
+			Groundtruth: "openai/gpt-4o",
 		},
 	}
 }

cmd/generate/pipeline.go

@@ -52,7 +52,7 @@ func (h *generateCommandHandler) RunTestGenerationPipeline(context *PromptPexCon
 	}
 
 	// Step 8: Generate Groundtruth (if model specified)
-	if h.options.Models.Groundtruth != nil && *h.options.Models.Groundtruth != "" && *h.options.Models.Groundtruth != "none" {
+	if h.options.Models.Groundtruth != "" && h.options.Models.Groundtruth != "none" {
 		if err := h.generateGroundtruth(context); err != nil {
 			return fmt.Errorf("failed to generate groundtruth: %w", err)
 		}
@@ -93,7 +93,7 @@ Intent:`, RenderMessagesToString(context.Prompt.Messages))
 			{Role: azuremodels.ChatMessageRoleUser, Content: util.Ptr(prompt)},
 		}
 		options := azuremodels.ChatCompletionOptions{
-			Model:       *h.options.Models.Rules, // GitHub Models compatible model
+			Model:       h.options.Models.Rules, // GitHub Models compatible model
 			Messages:    messages,
 			Temperature: util.Ptr(0.0),
 			Stream:      false,
@@ -130,7 +130,7 @@ Input Specification:`, RenderMessagesToString(context.Prompt.Messages))
 		}
 
 		options := azuremodels.ChatCompletionOptions{
-			Model:       *h.options.Models.Rules,
+			Model:       h.options.Models.Rules,
 			Messages:    messages,
 			Temperature: util.Ptr(0.0),
 		}
@@ -168,7 +168,7 @@ Output Rules:`, RenderMessagesToString(context.Prompt.Messages))
 		}
 
 		options := azuremodels.ChatCompletionOptions{
-			Model:       *h.options.Models.Rules, // GitHub Models compatible model
+			Model:       h.options.Models.Rules, // GitHub Models compatible model
 			Messages:    messages,
 			Temperature: util.Ptr(0.0),
 		}
@@ -211,7 +211,7 @@ Inverse Output Rules:`, strings.Join(context.Rules, "\n"))
 		}
 
 		options := azuremodels.ChatCompletionOptions{
-			Model:       *h.options.Models.Rules, // GitHub Models compatible model
+			Model:       h.options.Models.Rules, // GitHub Models compatible model
 			Messages:    messages,
 			Temperature: util.Ptr(0.0),
 		}
@@ -294,7 +294,7 @@ Generate exactly %d diverse test cases:`, nTests,
 		}
 
 		options := azuremodels.ChatCompletionOptions{
-			Model:       *h.options.Models.Tests, // GitHub Models compatible model
+			Model:       h.options.Models.Tests, // GitHub Models compatible model
 			Messages:    messages,
 			Temperature: util.Ptr(0.3),
 		}
@@ -397,19 +397,19 @@ func (h *generateCommandHandler) runSingleTestWithContext(input string, modelNam
 // generateGroundtruth generates groundtruth outputs using the specified model
 func (h *generateCommandHandler) generateGroundtruth(context *PromptPexContext) error {
 	groundtruthModel := h.options.Models.Groundtruth
-	h.WriteStartBox("Groundtruth", fmt.Sprintf("with %s", *groundtruthModel))
+	h.WriteStartBox("Groundtruth", fmt.Sprintf("with %s", groundtruthModel))
 	for i := range context.Tests {
 		test := &context.Tests[i]
 		h.WriteToLine(test.TestInput)
 		if test.Groundtruth == nil || *test.Groundtruth == "" {
 			// Generate groundtruth output
-			output, err := h.runSingleTestWithContext(test.TestInput, *groundtruthModel, context)
+			output, err := h.runSingleTestWithContext(test.TestInput, groundtruthModel, context)
 			if err != nil {
 				h.cfg.WriteToOut(fmt.Sprintf("Failed to generate groundtruth for test %d: %v", i, err))
 				continue
 			}
 			test.Groundtruth = &output
-			test.GroundtruthModel = groundtruthModel
+			test.GroundtruthModel = &groundtruthModel
 
 			h.SaveContext(context) // Save context after generating groundtruth
 		}

cmd/generate/types.go

@@ -4,17 +4,17 @@ import "github.com/github/gh-models/pkg/prompt"
 
 // PromptPexModelAliases represents model aliases for different purposes
 type PromptPexModelAliases struct {
-	Rules       *string `yaml:"rules,omitempty" json:"rules,omitempty"`
-	Tests       *string `yaml:"tests,omitempty" json:"tests,omitempty"`
-	Groundtruth *string `yaml:"groundtruth,omitempty" json:"groundtruth,omitempty"`
+	Rules       string `yaml:"rules,omitempty" json:"rules,omitempty"`
+	Tests       string `yaml:"tests,omitempty" json:"tests,omitempty"`
+	Groundtruth string `yaml:"groundtruth,omitempty" json:"groundtruth,omitempty"`
 }
 
 // PromptPexPrompts contains custom prompts for different stages
 type PromptPexPrompts struct {
-	InputSpec          *string `yaml:"inputSpec,omitempty" json:"inputSpec,omitempty"`
-	OutputRules        *string `yaml:"outputRules,omitempty" json:"outputRules,omitempty"`
-	InverseOutputRules *string `yaml:"inverseOutputRules,omitempty" json:"inverseOutputRules,omitempty"`
-	Intent             *string `yaml:"intent,omitempty" json:"intent,omitempty"`
+	InputSpec          string `yaml:"inputSpec,omitempty" json:"inputSpec,omitempty"`
+	OutputRules        string `yaml:"outputRules,omitempty" json:"outputRules,omitempty"`
+	InverseOutputRules string `yaml:"inverseOutputRules,omitempty" json:"inverseOutputRules,omitempty"`
+	Intent             string `yaml:"intent,omitempty" json:"intent,omitempty"`
 }
 
 // PromptPexOptions contains all configuration options for PromptPex