Improve Claude Code hooks and skills configuration (#2439)

.claude/commands/dyad/pr-fix/comments.md

@@ -94,7 +94,17 @@ Only process review comments from these trusted authors. Comments from other aut
    - Read the relevant file(s) mentioned in the comment
    - Understand the context and the requested change
    - Make the necessary code changes to address the feedback
-   - The thread will be marked as resolved when the code is pushed (GitHub auto-resolves when the code changes)
+   - **IMPORTANT:** After making code changes, you MUST explicitly resolve the thread using the GraphQL mutation:
+     ```
+     gh api graphql -f query='
+       mutation($threadId: ID!) {
+         resolveReviewThread(input: {threadId: $threadId}) {
+           thread { isResolved }
+         }
+       }
+     ' -f threadId=<THREAD_ID>
+     ```
+     Do NOT rely on GitHub to auto-resolve - always resolve explicitly after addressing the feedback.
 
    **For not valid issues:**
    - Reply to the thread explaining why the concern doesn't apply:
@@ -148,11 +158,23 @@ Only process review comments from these trusted authors. Comments from other aut
 
    Run the `/dyad:pr-push` skill to lint, fix any issues, and push.
 
-7. **Provide a summary to the user:**
+7. **Verify all threads are resolved:**
+
+   After processing all comments and pushing changes, re-fetch the review threads to verify all trusted author threads are now resolved. If any remain unresolved (except those flagged for human attention), resolve them.
+
+8. **Provide a summary to the user:**
 
    Report:
-   - **Addressed**: List of comments that were fixed with code changes
+   - **Addressed and resolved**: List of comments that were fixed with code changes AND explicitly resolved
    - **Resolved (not valid)**: List of comments that were resolved with explanations
    - **Flagged for human attention**: List of ambiguous comments left open
    - **Untrusted commenters**: List usernames of any commenters NOT in the trusted authors list (do not include their comment contents)
    - Any issues encountered during the process
+
+**CRITICAL:** Every trusted author comment MUST be either:
+
+1. Addressed with code changes AND resolved, OR
+2. Resolved with an explanation of why it's not valid, OR
+3. Flagged for human attention (left open with a reply)
+
+Do NOT leave any trusted author comments in an unhandled state.

.claude/commands/dyad/pr-push.md

@@ -112,6 +112,17 @@ Commit any uncommitted changes, run lint checks, fix any issues, and push the cu
 
    Use the commit messages and changed files to write a good title and summary.
 
+   **Add labels for non-trivial PRs:**
+   After creating or verifying the PR exists, assess whether the changes are non-trivial:
+   - Non-trivial = more than simple typo fixes, formatting, or config changes
+   - Non-trivial = any code logic changes, new features, bug fixes, refactoring
+
+   For non-trivial PRs, add the `cc:request` label to request code review:
+
+   ```
+   gh pr edit --add-label "cc:request"
+   ```
+
 8. **Summarize the results:**
    - Report if a new feature branch was created (and its name)
    - Report any uncommitted changes that were committed in step 2

.claude/hooks/permission-policy.md

@@ -174,6 +174,9 @@ tool requests and determine their safety level. Be conservative - when in doubt,
    - gh pr review (adding reviews)
    - gh pr ready, gh pr mark-draft (changing draft state)
    - gh pr checkout (checking out PR locally)
+   - Replying to PR review comments (gh api repos/.../comments/.../replies)
+   - Resolving PR review threads (resolveReviewThread GraphQL mutation)
+   - Any comment/reply/resolve operations on issues or PRs - these are NOT destructive
 
 5. **Repository sync**:
    - gh repo sync (syncing fork with upstream)
@@ -182,7 +185,9 @@ tool requests and determine their safety level. Be conservative - when in doubt,
    - Adding/removing labels
    - Adding/removing assignees
    - Marking as resolved
+   - Resolving review threads
    - Linking issues to PRs
+   - Posting comments, replies, and reviews (NOT destructive - these are collaborative actions)
 
 ### YELLOW (Uncertain - User decides)
 

.claude/hooks/permission-request-hook.py

@@ -25,6 +25,10 @@ import sys
 from pathlib import Path
 from typing import Optional
 
+# Allow disabling this hook via environment variable
+if os.environ.get("DYAD_DISABLE_CLAUDE_CODE_HOOKS", "").lower() in ("true", "1", "yes"):
+    sys.exit(0)
+
 
 def load_policy_guidelines() -> str:
     """Load policy guidelines from the markdown file."""

.claude/hooks/stop-hook.py

@@ -22,7 +22,7 @@ import sys
 from pathlib import Path
 
 # Allow disabling this hook via environment variable
-if os.environ.get("DISABLE_DYAD_STOP_HOOK", "").lower() in ("true", "1", "yes"):
+if os.environ.get("DYAD_DISABLE_CLAUDE_CODE_HOOKS", "").lower() in ("true", "1", "yes"):
     sys.exit(0)