From ttal
Triage PR review comments by categorizing them as actionable, false positive, or deferrable, fixing actionable items, and posting a status update.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ttal:triage [review-file-path][review-file-path]The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Triage PR review comments: assess each one, fix what's actionable, then post a status update.
Triage PR review comments: assess each one, fix what's actionable, then post a status update.
If the reviewer notification includes a file path (e.g., /tmp/ttal-review-XXXXX.md),
read it first — it contains the full review.
If no file path was provided, use the review content already present in context.
Read all review comments and categorize them.
For each comment, ask:
Don't dismiss issues because they're preexisting. "Out of scope" and "predates this PR" are not reasons to defer. Evaluate by worth + effort, not by origin.
Actionable (fix now)
Format: [FIX] <summary> — <why it matters>
False Positive (push back)
Format: [FALSE POSITIVE] <summary> — <why it's wrong, suggested response>
Deferrable (follow-up)
Format: [DEFER] <summary> — <why it can wait>
Beyond reviewer comments, scan the PR diff for:
Include these as actionable even if the reviewer didn't mention them.
Address all actionable items. Then move to Phase 3.
Show a status update with what's done, what's remaining, and what you're pushing back on.
# Check what's changed since review
git log --oneline origin/main..HEAD
git diff origin/main..HEAD
For each item, verify against code: search for implementations, check tests, confirm fixes.
Format:
## Triage Update
### Fixed
- [x] Item — *addressed in commit abc123*
- [x] Item — *implemented in `src/file.ts:42`*
### False Positive
- [ ] Item — *[reason, suggested response]*
### Deferred
- [ ] Item — *[reason, follow-up plan]*
### Remaining
- [ ] Item — *[what's still needed]*
npx claudepluginhub tta-lab/ttal-cli --plugin ttalTriages PR review comments by filtering false positives (outdated, misread, wrong context, style preference), validating against current diff, and fixing actionable ones with per-fix confirmation.
Loads GitHub PR review comments into the AI session for analysis, triage, and fix planning. Default is analysis-only; use --mode fix to enable auto-fixes.
Reads open GitHub PR review comments, triages by severity, applies code fixes, and drafts replies. Use when addressing PR feedback or code review.