Answer Reviewer Questions
Draft replies to code review questions using implementation reasoning.
Installation
- Make sure Claude is on your device and in your terminal.
Skills load from
~/.claude/skills/when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then runclaudein any terminal to verify.One-time setupnpm i -g @anthropic-ai/claude-codeAlready have it? Skip ahead.
- Paste into Claude Code or into your terminal.
This copies the whole skill folder into
~/.claude/skills/answer-reviewer-questions-tobihagemann/— the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.Faster alternative (instruction-only skills)
Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.
Quick install (SKILL.md only)Sign up to copy - Restart Claude Code.
Quit and reopen Claude Code (or any other agent that loads from
~/.claude/skills/). New skills are picked up on startup. - Just ask Claude.
Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.
Prefer to read the source first? Open on GitHub.
When Claude uses it
For each reviewer question on a PR, recall implementation reasoning and compose a raw answer. Use when the user asks to "answer reviewer questions", "draft answers to PR questions", or "explain reviewer questions".
What this skill does
Answer Reviewer Questions
For each reviewer question thread, recall the implementer's reasoning and compose a raw answer. The answers are plain text and feed into a downstream reply-drafting skill that applies voice rules and reply formatting.
Step 1: Collect Question Threads
Use the question threads from conversation context. Each thread has: thread id, file path, line (use originalLine when line is null), the reviewer's original comment, and the reconciled intent from /interpret-feedback.
If no question threads were provided, report that there are no questions to answer and stop.
Step 2: Answer Each Thread
For each thread:
- Run the
/recall-reasoningskill with<path>:<line>. It returns either recalled reasoning from a past transcript, or a fallback derived from reading the commit diff and surrounding code. - Compose a one-or-two-sentence answer from the returned reasoning. Quote or paraphrase the implementer's own words when the recalled reasoning explains the decision well.
- Do not mention Claude, transcripts, or that the reasoning was recalled. The answer reads as the implementer's own explanation.
Step 3: Output Answers
Output one block per thread:
**Thread <id>** (<path>:<line>)
<answer text>
_Grounding: derived from current code_
Include the _Grounding:_ line only when /recall-reasoning returned no transcript. Omit it when the answer is grounded in recalled reasoning.
Then use the TaskList tool and proceed to any remaining task.
Rules
- Do not load
/github-voiceor apply reply formatting. Downstream drafting applies voice rules when composing the actual reply. - When
/recall-reasoningreturned no transcript, still compose an answer from the current code and include the_Grounding:_line so the downstream drafter knows the answer has weaker grounding.
Related skills
Generative Code Art
anthropics
Create algorithmic art with p5.js using randomness and interactive parameters.
Poster & Visual Design
anthropics
Create original posters and visual art in PNG and PDF formats.
Claude API Helper
anthropics
Build, debug, and optimize Claude API applications with caching and model migration support.
MCP Server Builder
anthropics
Build protocol servers that connect language models to external APIs and services.