Simplify agent mode and re-add additional_permissions input

- Agent mode now only triggers when explicit prompt is provided
- Removed automatic triggering for workflow_dispatch/schedule without prompt
- Re-added additional_permissions input for requesting GitHub permissions
- Fixed TypeScript types for mock context helpers to properly handle partial inputs
- Updated documentation to reflect simplified mode behavior

src/modes/agent/index.ts

@@ -1,23 +1,22 @@
 import * as core from "@actions/core";
 import { mkdir, writeFile } from "fs/promises";
 import type { Mode, ModeOptions, ModeResult } from "../types";
-import { isAutomationContext } from "../../github/context";
 import type { PreparedContext } from "../../create-prompt/types";
 
 /**
  * Agent mode implementation.
  *
- * This mode is specifically designed for automation events (workflow_dispatch and schedule).
- * It bypasses the standard trigger checking and comment tracking used by tag mode,
- * making it ideal for scheduled tasks and manual workflow runs.
+ * This mode runs whenever an explicit prompt is provided in the workflow configuration.
+ * It bypasses the standard @claude mention checking and comment tracking used by tag mode,
+ * providing direct access to Claude Code for automation workflows.
  */
 export const agentMode: Mode = {
   name: "agent",
-  description: "Automation mode for workflow_dispatch and schedule events",
+  description: "Direct automation mode for explicit prompts",
 
   shouldTrigger(context) {
-    // Only trigger for automation events
-    return isAutomationContext(context);
+    // Only trigger when an explicit prompt is provided
+    return !!context.inputs?.prompt;
   },
 
   prepareContext(context) {
@@ -41,7 +40,7 @@ export const agentMode: Mode = {
   },
 
   async prepare({ context }: ModeOptions): Promise<ModeResult> {
-    // Agent mode handles automation events (workflow_dispatch, schedule) only
+    // Agent mode handles automation events and any event with explicit prompts
 
     // TODO: handle by createPrompt (similar to tag and review modes)
     // Create prompt directory

src/modes/detector.ts

@@ -9,13 +9,7 @@ import { checkContainsTrigger } from "../github/validation/trigger";
 export type AutoDetectedMode = "tag" | "agent";
 
 export function detectMode(context: GitHubContext): AutoDetectedMode {
-  // If prompt is provided, always use agent mode
-  // Reasoning: When users provide explicit instructions via the prompt parameter,
-  // they want Claude to execute those instructions immediately without waiting for
-  // @claude mentions or other triggers. This aligns with the v1.0 philosophy where
-  // Claude Code handles everything - the GitHub Action is just a thin wrapper that
-  // passes through prompts directly to Claude Code for native handling (including
-  // slash commands). This provides the most direct and flexible interaction model.
+  // If prompt is provided, use agent mode for direct execution
   if (context.inputs?.prompt) {
     return "agent";
   }
@@ -38,7 +32,7 @@ export function detectMode(context: GitHubContext): AutoDetectedMode {
     }
   }
 
-  // Default to agent mode for everything else
+  // Default to agent mode (which won't trigger without a prompt)
   return "agent";
 }
 
@@ -47,7 +41,7 @@ export function getModeDescription(mode: AutoDetectedMode): string {
     case "tag":
       return "Interactive mode triggered by @claude mentions";
     case "agent":
-      return "General automation mode for all events";
+      return "Direct automation mode for explicit prompts";
     default:
       return "Unknown mode";
   }

src/modes/types.ts

@@ -26,7 +26,7 @@ export type ModeData = {
  *
  * Current modes include:
  * - 'tag': Interactive mode triggered by @claude mentions
- * - 'agent': General automation mode for all events (default)
+ * - 'agent': Direct automation mode triggered by explicit prompts
  */
 export type Mode = {
   name: ModeName;