feat: add agent mode for automation scenarios

- Add agent mode that always triggers without checking for mentions
- Implement Mode interface with support for mode-specific tool configuration
- Add getAllowedTools() and getDisallowedTools() methods to Mode interface
- Simplify tests by combining related test cases
- Update documentation and examples to include agent mode
- Fix TypeScript imports to prevent circular dependencies

Agent mode is designed for automation and workflow_dispatch scenarios
where Claude should always run without requiring trigger phrases.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

examples/claude-modes.yml

@@ -0,0 +1,117 @@
+name: Claude Code Action Mode Examples
+
+on:
+  # Example 1: Tag mode (default) - traditional implementation triggered by mentions
+  issue_comment:
+    types: [created]
+  issues:
+    types: [opened, assigned, labeled]
+  pull_request:
+    types: [opened, synchronize]
+  pull_request_review_comment:
+    types: [created]
+  pull_request_review:
+    types: [submitted]
+
+  # Example 2: Agent mode - for scheduled tasks and automation
+  schedule:
+    - cron: "0 0 * * 1" # Weekly on Monday
+  workflow_dispatch:
+    inputs:
+      task:
+        description: "Task for Claude to perform"
+        required: true
+        type: string
+
+jobs:
+  # Example 1: Tag mode (default) - triggered by @claude mentions
+  claude-tag-mode:
+    if: github.event_name != 'schedule' && github.event_name != 'workflow_dispatch'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: anthropics/claude-code-action@main
+        with:
+          # mode defaults to 'tag' - no need to specify
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          # Responds to @claude mentions, assignments, and labels
+          # Creates tracking comments showing progress
+          # Full implementation capabilities
+
+  # Example 2: Agent mode - for automation
+  claude-agent-scheduled:
+    if: github.event_name == 'schedule'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: anthropics/claude-code-action@main
+        with:
+          mode: agent
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          override_prompt: |
+            Check the repository for:
+            1. Outdated dependencies in package.json
+            2. TODO comments that haven't been addressed
+            3. Files that haven't been updated in over 6 months
+
+            Create an issue summarizing your findings.
+          # Always runs - no trigger checking
+          # Perfect for scheduled tasks
+
+  # Example 3: Agent mode with workflow_dispatch
+  claude-agent-dispatch:
+    if: github.event_name == 'workflow_dispatch'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: anthropics/claude-code-action@main
+        with:
+          mode: agent
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          override_prompt: |
+            Task: ${{ github.event.inputs.task }}
+
+            Repository: $REPOSITORY
+            Triggered by: $TRIGGER_USERNAME
+
+            Please complete the requested task and provide a summary of what was done.
+
+  # Example 4: Custom security review with override_prompt
+  claude-security-review:
+    if: github.event.pull_request.draft == false && contains(github.event.pull_request.labels.*.name, 'security-review')
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: anthropics/claude-code-action@main
+        with:
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          override_prompt: |
+            Perform a security review of PR #$PR_NUMBER.
+
+            Focus on:
+            - SQL injection vulnerabilities
+            - XSS (Cross-Site Scripting) risks
+            - Authentication/authorization issues
+            - Sensitive data exposure
+            - Input validation problems
+
+            Changed files:
+            $CHANGED_FILES
+
+            Provide severity ratings (Critical/High/Medium/Low) for any issues found.
+            Be specific about file paths and line numbers.