chore(workflows): update model references to claude-opus-5 (#558)

Anthropic released Claude Opus 5 (claude-opus-5), which supersedes
claude-opus-4-8 as the latest Opus model. Per the model-watch job,
`next` still referenced claude-opus-4-8 in 19 files even though the
prior model-update pass (#548) already moved Sonnet to claude-sonnet-5.

Like-for-like ID/prose swap only, pricing unchanged ($5/$25 per 1M
tokens for both). claude-sonnet-5 and claude-haiku-4-5 references are
untouched (already current). Two files documenting Boris Cherny's
personal model preference (claude-setup/README.md and
boris-best-practices.md) are intentionally left as-is since they quote
him by name; rewriting his stated preference would misrepresent the
source rather than update a config value.

Bumped development-codebase-tools (2.6.2 -> 2.6.3) and
development-planning (2.0.7 -> 2.0.8) plugin.json versions per this
repo's mandatory version-bump-on-plugin-file-change convention.

Author: wkoutre

.github/REUSABLE_WORKFLOWS.md

@@ -218,7 +218,7 @@ Notify Release (_notify-release.yml)
 
 | Input                           | Required | Default                          | Description                                                                        |
 | ------------------------------- | -------- | -------------------------------- | ---------------------------------------------------------------------------------- |
-| `model`                         | No       | `'claude-sonnet-5'`              | Claude model to use (Sonnet 5, Opus 4.8, or Haiku 4.5)                             |
+| `model`                         | No       | `'claude-sonnet-5'`              | Claude model to use (Sonnet 5, Opus 5, or Haiku 4.5)                               |
 | `allowed_tools`                 | No       | (permissive defaults, see below) | YAML string specifying which tools Claude can use (file operations, bash commands) |
 | `custom_instructions`           | No       | `'Be sure to follow rules...'`   | Additional instructions for Claude beyond CLAUDE.md files                          |
 | `timeout_minutes`               | No       | `'10'`                           | Maximum execution time in minutes (prevents runaway costs)                         |
@@ -282,7 +282,7 @@ The following settings are intentionally fixed to ensure consistent security and
 
 - **Interactive AI Assistance**: Respond to `@claude` mentions anywhere in GitHub
 - **Multiple Trigger Points**: Works in issue comments, PR comments, review comments, and reviews
-- **Configurable Model**: Choose between Sonnet 5 (balanced), Opus 4.8 (thorough), or Haiku 4.5 (fast)
+- **Configurable Model**: Choose between Sonnet 5 (balanced), Opus 5 (thorough), or Haiku 4.5 (fast)
 - **Flexible Tool Permissions**: Control what Claude can do (read-only, read-write, or custom)
 - **Custom Instructions**: Add repository-specific guidelines and standards
 - **Bot Filtering**: Automatically excludes bot comments to prevent loops
@@ -334,7 +334,7 @@ jobs:
     secrets:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
     with:
-      model: 'claude-opus-4-8'
+      model: 'claude-opus-5'
       timeout_minutes: '15' # Opus may need more time
 ```
 
@@ -423,7 +423,7 @@ jobs:
     secrets:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
     with:
-      model: 'claude-opus-4-8'
+      model: 'claude-opus-5'
       timeout_minutes: '15'
 ```
 
@@ -546,7 +546,7 @@ Claude says it doesn't have permission to perform action
 2. **Choose the Right Model**:
 
    - **Sonnet 5** (default): Best balance of speed, capability, and cost for 90% of use cases
-   - **Opus 4.8**: Reserve for complex architectural reviews or critical security analysis
+   - **Opus 5**: Reserve for complex architectural reviews or critical security analysis
    - **Haiku 4.5**: Fast and cost-effective for simple questions, quick lookups, or high-volume usage
 
 3. **Set Appropriate Timeouts**:
@@ -619,7 +619,7 @@ jobs:
     secrets:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
     with:
-      model: 'claude-opus-4-8'
+      model: 'claude-opus-5'
       timeout_minutes: '15'
 ```
 
@@ -640,7 +640,7 @@ claude-deep:
     contains(github.event.pull_request.labels.*.name, 'deep-analysis') &&
     # ... (rest of if condition)
   with:
-    model: 'claude-opus-4-8'
+    model: 'claude-opus-5'
 ```
 
 #### Integration with Other Workflows
@@ -691,7 +691,7 @@ Tips for managing Claude API costs effectively:
 
    - Haiku 4.5: Most cost-effective for simple tasks
    - Sonnet 5: ~$2 per 1M input tokens, ~$10 per 1M output tokens (introductory pricing through Aug 31, 2026; default, recommended)
-   - Opus 4.8: ~$5 per 1M input tokens, ~$25 per 1M output tokens (reserve for complex tasks)
+   - Opus 5: ~$5 per 1M input tokens, ~$25 per 1M output tokens (reserve for complex tasks)
    - Typical interaction: 5K-50K tokens (mostly input)
 
 2. **Timeout Strategy**:
@@ -1117,7 +1117,7 @@ jobs:
       pr_number: ${{ github.event.pull_request.number }}
       base_ref: ${{ github.base_ref }}
       # Use Opus for PRs with 'claude-opus' label, otherwise Sonnet
-      model: ${{ contains(github.event.pull_request.labels.*.name, 'claude-opus') && 'claude-opus-4-8' || 'claude-sonnet-5' }}
+      model: ${{ contains(github.event.pull_request.labels.*.name, 'claude-opus') && 'claude-opus-5' || 'claude-sonnet-5' }}
       max_turns: 20 # Allow more turns for thorough Opus reviews
       timeout_minutes: 45 # Longer timeout for complex reviews
     secrets:
@@ -1501,7 +1501,7 @@ Claude submitted multiple reviews instead of updating one
 2. **Model Selection**:
 
    - **Sonnet 5** (default): Best balance for most PRs (~80% of use cases)
-   - **Opus 4.8**: Reserve for critical code, security-sensitive changes, or complex architectures
+   - **Opus 5**: Reserve for critical code, security-sensitive changes, or complex architectures
    - Use labels to let developers request deeper reviews when needed
 
 3. **Custom Prompts**:
@@ -1596,7 +1596,7 @@ jobs:
     with:
       pr_number: ${{ github.event.pull_request.number }}
       base_ref: ${{ github.base_ref }}
-      model: 'claude-opus-4-8'
+      model: 'claude-opus-5'
       custom_prompt_path: '.github/prompts/security-review.md'
       timeout_minutes: 60
     secrets:

.github/workflows/CLAUDE.md

@@ -435,7 +435,7 @@ with:
   pr_number: ${{ github.event.pull_request.number }}
   base_ref: ${{ github.base_ref }}
   auto_fix: true # Enable automatic fixing of issues
-  auto_fix_model: 'claude-opus-4-8' # Use Opus for better fixes (optional)
+  auto_fix_model: 'claude-opus-5' # Use Opus for better fixes (optional)
 secrets:
   ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
   WORKFLOW_PAT: ${{ secrets.WORKFLOW_PAT }} # Required for pushing fixes
@@ -745,7 +745,7 @@ uses: Uniswap/ai-toolkit/.github/workflows/_claude-docs-check.yml@main
 with:
   pr_number: ${{ github.event.pull_request.number }}
   auto_fix: true # Enable automatic fixing of documentation issues
-  auto_fix_model: 'claude-opus-4-8' # Use Opus for better fixes (optional)
+  auto_fix_model: 'claude-opus-5' # Use Opus for better fixes (optional)
 secrets:
   ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
   WORKFLOW_PAT: ${{ secrets.WORKFLOW_PAT }} # Required for pushing fixes
@@ -1055,14 +1055,14 @@ If both are provided, OAuth token takes precedence. At least one authentication
 
 **Configuration:**
 
-| Input                     | Default           | Description                                     |
-| ------------------------- | ----------------- | ----------------------------------------------- |
-| `model`                   | `claude-opus-4-8` | Claude model to use                             |
-| `max_turns`               | `150`             | Maximum conversation turns                      |
-| `debug_mode`              | `true`            | Show full Claude output                         |
-| `timeout_minutes`         | `60`              | Job timeout                                     |
-| `pr_type`                 | `draft`           | Type of PR to create: "draft" or "published"    |
-| `install_uniswap_plugins` | `true`            | Auto-install uniswap plugins (false to opt out) |
+| Input                     | Default         | Description                                     |
+| ------------------------- | --------------- | ----------------------------------------------- |
+| `model`                   | `claude-opus-5` | Claude model to use                             |
+| `max_turns`               | `150`           | Maximum conversation turns                      |
+| `debug_mode`              | `true`          | Show full Claude output                         |
+| `timeout_minutes`         | `60`            | Job timeout                                     |
+| `pr_type`                 | `draft`         | Type of PR to create: "draft" or "published"    |
+| `install_uniswap_plugins` | `true`          | Auto-install uniswap plugins (false to opt out) |
 
 **Validation Behavior:**
 
@@ -1096,7 +1096,7 @@ with:
   issue_url: ${{ matrix.issue_url }}
   branch_name: ${{ matrix.branch_name }}
   target_branch: 'next'
-  model: 'claude-opus-4-8'
+  model: 'claude-opus-5'
   debug_mode: true
   pr_type: 'draft' # or 'published' for non-draft PRs
 secrets:
@@ -1116,7 +1116,7 @@ with:
   issue_url: ${{ matrix.issue_url }}
   branch_name: ${{ matrix.branch_name }}
   target_branch: 'next'
-  model: 'claude-opus-4-8'
+  model: 'claude-opus-5'
   debug_mode: true
   pr_type: 'draft'
 secrets:
@@ -1208,7 +1208,7 @@ gh workflow run update-action-versions.yml
 gh workflow run update-action-versions.yml -f dry_run=true
 
 # Use Opus model
-gh workflow run update-action-versions.yml -f model=claude-opus-4-8
+gh workflow run update-action-versions.yml -f model=claude-opus-5
 ```
 
 **Usage example (API Key):**
@@ -1344,7 +1344,7 @@ gh workflow run dev-ai-newsletter.yml -f dry_run=true
 gh workflow run dev-ai-newsletter.yml -f days_back=14
 
 # Use Opus model for better quality
-gh workflow run dev-ai-newsletter.yml -f model=claude-opus-4-8
+gh workflow run dev-ai-newsletter.yml -f model=claude-opus-5
 
 # Post to specific Slack channels
 gh workflow run dev-ai-newsletter.yml -f slack_post_channel_ids="C091XE1DNP2,C094URH6C13"

.github/workflows/README.md

@@ -75,7 +75,7 @@ Workflows designed to be called by other workflows using `workflow_call`. These
 
   - Interactive AI assistance via @claude mentions
   - Works in issue comments, PR comments, review comments, and reviews
-  - Configurable models (Sonnet 5, Opus 4.8, Haiku 4.5)
+  - Configurable models (Sonnet 5, Opus 5, Haiku 4.5)
   - Flexible tool permissions (read-only, read-write, or custom)
   - Custom instructions for repository-specific standards
   - Fixed security settings (Bullfrog scanning, immutable permissions)

.github/workflows/_claude-main.yml

@@ -79,7 +79,7 @@ on:
 
           Available models:
           - claude-sonnet-5 (default, recommended) - Latest Sonnet model with best balance of speed and capability
-          - claude-opus-4-8 - Most capable model for complex tasks
+          - claude-opus-5 - Most capable model for complex tasks
           - claude-haiku-4-5-20251001 - Fast, cost-effective model for simpler tasks
 
           Default: claude-sonnet-5

.github/workflows/_claude-task-worker.yml

@@ -89,10 +89,10 @@ on:
       model:
         description: |
           Claude model to use.
-          Options: claude-sonnet-5, claude-opus-4-8, claude-haiku-4-5-20251001
+          Options: claude-sonnet-5, claude-opus-5, claude-haiku-4-5-20251001
         required: false
         type: string
-        default: "claude-opus-4-8"
+        default: "claude-opus-5"
 
       timeout_minutes:
         description: "Maximum execution time in minutes"

.github/workflows/_generate-pr-metadata.yml

@@ -47,7 +47,7 @@ on:
         description: "Claude model to use for generation"
         required: false
         type: string
-        default: "claude-opus-4-8"
+        default: "claude-opus-5"
 
       max_turns:
         description: "Maximum conversation turns for Claude. Omit to use unlimited turns."

.github/workflows/claude-auto-tasks.yml

@@ -53,9 +53,9 @@ on:
         type: choice
         options:
           - "claude-sonnet-5"
-          - "claude-opus-4-8"
+          - "claude-opus-5"
           - "claude-haiku-4-5"
-        default: "claude-opus-4-8"
+        default: "claude-opus-5"
 
       target_branch:
         description: "Branch to create PRs against"
@@ -124,7 +124,7 @@ jobs:
       issue_url: ${{ matrix.issue_url }}
       branch_name: ${{ matrix.branch_name }}
       target_branch: ${{ inputs.target_branch || 'next' }}
-      model: ${{ inputs.model || 'claude-opus-4-8' }}
+      model: ${{ inputs.model || 'claude-opus-5' }}
       timeout_minutes: 60
       linear_task_utils_tag: ${{ needs.prepare.outputs.linear_task_utils_tag }}
       debug_mode: ${{ inputs.debug_mode != '' && inputs.debug_mode || true }}

.github/workflows/dev-ai-newsletter.yml

@@ -97,7 +97,7 @@ on:
         type: choice
         options:
           - "claude-sonnet-5"
-          - "claude-opus-4-8"
+          - "claude-opus-5"
         default: "claude-sonnet-5"
 
       dry_run:

.github/workflows/examples/06-claude-main-basic.yml

@@ -92,7 +92,7 @@ jobs:
 #     secrets:
 #       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
 #     with:
-#       model: 'claude-opus-4-8'  # More capable, slower, higher cost
+#       model: 'claude-opus-5'  # More capable, slower, higher cost
 #       timeout_minutes: '15'  # Opus may need more time
 #
 # -------------------
@@ -204,7 +204,7 @@ jobs:
 #     secrets:
 #       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
 #     with:
-#       model: 'claude-opus-4-8'
+#       model: 'claude-opus-5'
 #       timeout_minutes: '15'
 #
 

.github/workflows/examples/07-claude-main-custom.yml

@@ -48,9 +48,9 @@ jobs:
       # MODEL SELECTION
       # Choose the Claude model based on your needs:
       # - claude-sonnet-5: Latest, best balance (RECOMMENDED)
-      # - claude-opus-4-8: Most capable, slower, higher cost
+      # - claude-opus-5: Most capable, slower, higher cost
       # - claude-haiku-4-5: Fast, cost-effective for simpler tasks
-      model: "claude-opus-4-8"
+      model: "claude-opus-5"
 
       # TIMEOUT CONFIGURATION
       # Set maximum execution time in minutes
@@ -256,7 +256,7 @@ jobs:
 #     secrets:
 #       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
 #     with:
-#       model: 'claude-opus-4-8'  # Thorough for code review
+#       model: 'claude-opus-5'  # Thorough for code review
 #       timeout_minutes: '15'
 #
 # -------------------
@@ -289,7 +289,7 @@ jobs:
 #     secrets:
 #       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
 #     with:
-#       model: 'claude-opus-4-8'
+#       model: 'claude-opus-5'
 #       timeout_minutes: '30'
 #       custom_instructions: |
 #         Perform a comprehensive codebase audit focusing on:
@@ -368,7 +368,7 @@ jobs:
 # Solution:
 #   1. Verify model name exactly matches available models:
 #      - claude-sonnet-5
-#      - claude-opus-4-8
+#      - claude-opus-5
 #      - claude-haiku-4-5
 #   2. Check Anthropic API key has access to requested model
 #   3. Review Anthropic console for API limits or restrictions