diff --git a/cmd/gh-aw/argument_syntax_test.go b/cmd/gh-aw/argument_syntax_test.go
index 38d732a835..46a53f957f 100644
--- a/cmd/gh-aw/argument_syntax_test.go
+++ b/cmd/gh-aw/argument_syntax_test.go
@@ -24,9 +24,9 @@ func TestArgumentSyntaxConsistency(t *testing.T) {
 	}{
 		// Commands with required arguments (using angle brackets <>)
 		{
-			name:           "audit command requires run-id",
+			name:           "audit command requires run-id-or-url",
 			command:        cli.NewAuditCommand(),
-			expectedUse:    "audit <run-id>",
+			expectedUse:    "audit <run-id-or-url>",
 			argsValidator:  "ExactArgs(1)",
 			shouldValidate: func(cmd *cobra.Command) error { return cmd.Args(cmd, []string{"123456"}) },
 		},
diff --git a/pkg/cli/add_command.go b/pkg/cli/add_command.go
index 1c7305c933..6b2020526e 100644
--- a/pkg/cli/add_command.go
+++ b/pkg/cli/add_command.go
@@ -74,7 +74,7 @@ Workflow specifications:
 
 The -n flag allows you to specify a custom name for the workflow file (only applies to the first workflow when adding multiple).
 The --dir flag allows you to specify a subdirectory under .github/workflows/ where the workflow will be added.
-The --create-pull-request flag (or --pr) creates a pull request with the workflow changes.
+The --create-pull-request flag creates a pull request with the workflow changes.
 The --force flag overwrites existing workflow files.
 
 Note: To create a new workflow from scratch, use the 'new' command instead.
diff --git a/pkg/cli/audit.go b/pkg/cli/audit.go
index bfe3d96d0b..6fca34096f 100644
--- a/pkg/cli/audit.go
+++ b/pkg/cli/audit.go
@@ -26,8 +26,8 @@ var auditLog = logger.New("cli:audit")
 // NewAuditCommand creates the audit command
 func NewAuditCommand() *cobra.Command {
 	cmd := &cobra.Command{
-		Use:   "audit <run-id>",
-		Short: "Audit a workflow run and generate a detailed report",
+		Use:   "audit <run-id-or-url>",
+		Short: "Audit a workflow run given a run ID or URL and generate a detailed report",
 		Long: `Audit a single workflow run by downloading artifacts and logs, detecting errors,
 analyzing MCP tool usage, and generating a concise Markdown report suitable for AI agents.
 
diff --git a/pkg/cli/mcp_inspect.go b/pkg/cli/mcp_inspect.go
index 68b0021c55..27eef34910 100644
--- a/pkg/cli/mcp_inspect.go
+++ b/pkg/cli/mcp_inspect.go
@@ -201,7 +201,7 @@ The command will:
 - Start each MCP server (stdio, docker, http)
 - Automatically start and inspect mcp-scripts server if present
 - Query available tools, resources, and roots
-- Validate required secrets are available  
+- Validate required secrets are available
 - Display results in formatted tables with error details`,
 		Args: cobra.MaximumNArgs(1),
 		RunE: func(cmd *cobra.Command, args []string) error {
diff --git a/pkg/cli/trial_command.go b/pkg/cli/trial_command.go
index f0c8fd3c0d..97141023d7 100644
--- a/pkg/cli/trial_command.go
+++ b/pkg/cli/trial_command.go
@@ -90,7 +90,7 @@ Workflows from different repositories:
   ` + string(constants.CLIExtensionPrefix) + ` trial githubnext/agentics/daily-plan myorg/myrepo/custom-workflow
 
 Repository mode examples:
-  ` + string(constants.CLIExtensionPrefix) + ` trial githubnext/agentics/my-workflow --repo myorg/myrepo            # Run directly in myorg/myrepo (no simulation)
+  ` + string(constants.CLIExtensionPrefix) + ` trial githubnext/agentics/my-workflow --host-repo myorg/myrepo         # Run directly in myorg/myrepo (no simulation)
   ` + string(constants.CLIExtensionPrefix) + ` trial githubnext/agentics/my-workflow --logical-repo myorg/myrepo  # Simulate running against myorg/myrepo
   ` + string(constants.CLIExtensionPrefix) + ` trial githubnext/agentics/my-workflow --clone-repo myorg/myrepo   # Clone myorg/myrepo contents into host
 
@@ -110,7 +110,7 @@ Advanced examples:
 Repository modes:
 - Default mode (no flags): Creates a temporary trial repository and simulates execution as if running against the current repository (github.repository context points to current repo)
 - --logical-repo REPO: Simulates execution against a specified repository (github.repository context points to REPO while actually running in a temporary trial repository)
-- --host-repo REPO (or --repo REPO): Uses the specified repository as the host for trial execution instead of creating a temporary one
+- --host-repo REPO: Uses the specified repository as the host for trial execution instead of creating a temporary one
 - --clone-repo REPO: Clones the specified repository's contents into the trial repository before execution (useful for testing against actual repository state)
 
 All workflows must support workflow_dispatch trigger to be used in trial mode.
@@ -182,6 +182,7 @@ Trial results are saved both locally (in trials/ directory) and in the host repo
 
 	cmd.Flags().String("host-repo", "", "Custom host repository slug (defaults to '<username>/gh-aw-trial'). Use '.' for current repository")
 	cmd.Flags().String("repo", "", "Alias for --host-repo: the repository where workflows are installed and run (note: different semantics from --repo in other commands)")
+	_ = cmd.Flags().MarkHidden("repo") // Hide alias to avoid semantic confusion with --repo in other commands
 	cmd.Flags().Bool("delete-host-repo-after", false, "Delete the host repository after completion (default: keep)")
 	cmd.Flags().Bool("force-delete-host-repo-before", false, "Force delete the host repository before creation, if it exists before creating it")
 	cmd.Flags().BoolP("yes", "y", false, "Skip confirmation prompts")