pr-comments-address
Use when the user wants a PR's review comments handled — triggers on "/pr-comments-address", "address the review comments", "handle the reviewer feedback"…
Skill by Esequiel378
Chalk Labs
Need help deploying pr-comments-address?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/esequiel378-pr-comments-address
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, sales
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/pr-comments-address
What's inside
- Skill name
pr-comments-address- Description
- Use when the user wants a PR's review comments handled — triggers on "/pr-comments-address", "address the review comments", "handle the reviewer feedback", "apply the review suggestions", or pointing at a PR with unresolved comments and asking to deal with them. Output is local working-tree changes plus a per-comment verdict report — never commits, pushes, or replies.
- License
- MIT
- Instructions
- 106 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Address PR Comments
Reviewer comments in, verdicts and local fixes out. Every comment gets judged —
not obeyed — and the accepted ones get fixed in the working tree. **This skill
never writes to the PR.** Committing, pushing, and replying are the user's
moves; the deliverable here is a dirty tree and a report they can act on.
## 0. Resolve the PR, fetch every unresolved comment
- **PR number / URL given** → use it. **Nothing given** → the current branch's
PR: `gh pr view --json number,headRefOid,baseRefName,url`.
- **Is the local branch what the PR shows?** Compare `git rev-parse HEAD` to the
PR's `headRefOid`. Comments anchor to lines in the PR's head; a diverged local
branch means fixing code the reviewer never commented on — stop and say so.
- Fetch all three comment surfaces (each misses the others):
- **Inline threads, with resolution state** — REST can't tell you what's
resolved, so use GraphQL:
```bash
gh api graphql -f query='query($o:String!,$r:String!,$n:Int!){
repository(owner:$o,name:$r){pullRequest(number:$n){
reviewThreads(first:100){nodes{isResolved isOutdated path line
comments(first:20){nodes{author{login} body}}}}}}}' \
-f o=<owner> -f r=<repo> -F n=<pr>
```
- **Review bodies** — `gh api repos/<o>/<r>/pulls/<n>/reviews`
- **Conversation comments** — `gh api repos/<o>/<r>/issues/<n>/comments`
- Work the **unresolved** threads; skip resolved ones and bot chatter unless the
user says otherwise. Outdated-but-unresolved threads still count — check
whether the concern survived the code that outdated them.
## 1. Triage — every comment gets exactly one verdict
Read the code each comment anchors to — the file as it is, not just the diff
hunk; the surrounding lines are where wrong verdicts come from. Then assign:
| Verdict | Meaning |
|---|---|
| **accept** | Valid, and this PR's job → gets fixed in step 3 |
| **descope** | Valid, but not this change's job → name the follow-up it belongs in |
| **move** | Valid, but belongs to a different PR → name which one |
| **reject** | Wrong or moot → cite the evidence (code, test, doc) that says so |
- **Stacks:** if the branch is part of a stack (use the `gh-stack` skill;
`gh stack view` maps the layers), a comment left on this PR about code
introduced in a *parent* layer is a **move** — fixing it here buries the fix
in the wrong diff and dooms the restack. Name the layer it belongs to.
- **Scope:** "while you're here, also…" comments that widen the PR beyond its
stated intent default to **descope**. A PR that grows in review is a PR that
never merges.
- **reject needs receipts.** A reviewer is a human who read the code; overruling
them on vibes is how valid feedback gets lost. No evidence → the verdict is
accept or descope, not reject.
## 2. Cross-validate contested verdicts with pr-review
Before finalizing any **reject**, and for any **accept** whose fix is invasive,
run the [pr-review](../pr-review/SKILL.md) skill scoped to the code in question.
The lenses are the second opinion: one independently re-raising the reviewer's
point flips a reject to accept; all of them passing the code silently is the
evidence a rejection can cite. Never reject a human's comment on your own
authority alone.
## 3. Fix the accepted ones — working tree only
- Fix the **root cause**, not just the line the comment names — grep the
sibling callers; a comment about one call site is usually about all of them.
- Run the tests and report the actual output — not "should still pass".
- Leave everything **uncommitted**. A dirty tree is the deliverable.
## 4. Report
```
## comments — PR #<n> (<addressed>/<total> unresolved threads)
| comment (file:line — author) | verdict | why / what changed |
|---|---|---|
| ... | accept | fixed in <file:line>: <one line> |
| ... | descope | belongs in <follow-up>: <one line> |
| ... | move | belongs in PR #<m> / stack layer <branch> |
| ... | reject | <evidence> |
Tests: <command> → <actual result>
Next moves are yours: review the diff, commit, push, reply.
```
Every unresolved comment appears in the table — a comment silently skipped
reads as a comment addressed.
## Guardrails
**Forbidden — no exceptions, regardless of urgency or convenience:**
- `git commit`, `git push` — in any form
- `gh pr comment`, `gh pr review`, reply endpoints, resolving threads,
re-requesting reviewers, reactions — anything that speaks on the PR
| Excuse | Reality |
|---|---|
| "Reviewers are waiting — I'll just push so they see it" | The user reviews the fixes before anyone else does. Speed pressure is a reason to triage faster, not to publish. |
| "One quick reply so they know it's handled" | A reply speaks *as the user* without their consent. Forbidden even as a courtesy. |
| "A local commit is harmless" | The user may want to reshape the branch history. The dirty tree *is* the output. |
- **Report what you skipped** — a thread you couldn't parse, a file too large, a
test you couldn't run. Silent partial coverage reads as full coverage.From Esequiel378/skills, MIT.
More from Esequiel378/skills
- 0.565score
pr-describe
Esequiel378
Use when writing or reviewing the body of a pull request — after `gh pr create`, when filling in an empty PR description, or when asked whether a PR body says…
SkillAFreshness: AClaude Code +9 - 0.565score
release-scaffold
Esequiel378
Use when a project needs a release process, versioned downloads, or a "Release" button — triggers on "/release-scaffold", "add a release process", "set up…
SkillAFreshness: AClaude Code +9 - 0.561score
architecture-improve
Esequiel378
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
SkillAFreshness: AClaude Code +9 - 0.539score
pr-review
Esequiel378
Use when the user wants a diff or PR reviewed from every angle at once — triggers on "/pr-review", "review this PR", "review my changes", "full review", or…
SkillAFreshness: AClaude Code +9 - 0.539score
codebase-design
Esequiel378
Shared vocabulary for designing deep modules.
SkillAFreshness: AClaude Code +9 - 0.530score
retro
Esequiel378
Conduct a retrospective on a coding session.
SkillAFreshness: AClaude Code +9 - 0.524score
prompt-improve
Esequiel378
Use when the user hands you a rough, vague, or messy prompt and wants it sharpened before acting — triggers on "improve this prompt", "fix my prompt"…
SkillBFreshness: BClaude Code +9 - 0.524score
prompt-improve-review
Esequiel378
Use to run the periodic analysis + tuning pass over the prompt-improve run log — triggers on "/prompt-improve-review", "review my prompt log", "tune…
SkillBFreshness: BClaude Code +9 - 0.524score
pr-review-sl
Esequiel378
Use when the user asks for a PR or diff review in the "sl" style — "sl review", "/pr-review-sl", "review this like sl" — a terse staff-architect pass that…
SkillBFreshness: BClaude Code +9 - 0.524score
commit-write
Esequiel378
Use when writing a commit message or a PR title — before `git commit`, before `gh pr create`, when renaming an existing PR, or when checking whether a commit…
SkillBFreshness: BClaude Code +9 - 0.524score
domain-modeling
Esequiel378
Use when building or sharpening a project's domain model — triggers on "/domain-modeling", discussing codebase terminology, naming a new concept, resolving…
SkillBFreshness: BClaude Code +9 - 0.524score
plan-map
Esequiel378
Use when a chunk of work is too big to hold in one agent session and the way to the end isn't visible yet — triggers on "/plan-map", "chart a map", "this is…
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 21, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Esequiel378/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Esequiel378
- 0.565score
release-scaffold
Esequiel378
Use when a project needs a release process, versioned downloads, or a "Release" button — triggers on "/release-scaffold", "add a release process", "set up…
SkillAFreshness: AClaude Code +9 - 0.565score
pr-describe
Esequiel378
Use when writing or reviewing the body of a pull request — after `gh pr create`, when filling in an empty PR description, or when asked whether a PR body says…
SkillAFreshness: AClaude Code +9 - 0.561score
architecture-improve
Esequiel378
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
SkillAFreshness: AClaude Code +9 - 0.539score
codebase-design
Esequiel378
Shared vocabulary for designing deep modules.
SkillAFreshness: AClaude Code +9 - 0.539score
pr-review
Esequiel378
Use when the user wants a diff or PR reviewed from every angle at once — triggers on "/pr-review", "review this PR", "review my changes", "full review", or…
SkillAFreshness: AClaude Code +9 - 0.530score
retro
Esequiel378
Conduct a retrospective on a coding session.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get pr-comments-address running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
pr-comments-address skill. Agent Mart. https://agentmart.dev/skills/esequiel378-pr-comments-address (last updated 2026-10-24)