rebuttal-guidance
Provides actionable peer-review rebuttal guidance (not full rebuttal prose) using conditional persuasion P(ZX,Y): match reviewer mindset X, recommend strategy Y, optionally ground in manuscript evidence and similar successful historical (review, rebuttal) pairs. Use when the user pastes reviewer comments, asks for rebuttal help, rebuttal guidance, response to reviewers, how to reply to R1.W2, OpenReview rebuttal, or the reviewer ignored the rebuttal.
- Installs
- 326
- Last 24h
- +9
- Trending rank
- #9282
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/hkustdial/supervisor-skills --skill rebuttal-guidance- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads rebuttal-guidance automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Rebuttal Guidance
Overview
This skill delivers executable rebuttal guidance per concern: structured plans, evidence pointers, tone and strategy priorities. It does not produce ready-to-paste rebuttal prose; the author writes the final text.
The core model is conditional persuasion P(Z=1 | X, Y):
- X: reviewer mindset (behavior cluster inferred from review text)
- Y: rebuttal strategy vector (which tactics to emphasize)
- Z: success (score increase or paper accepted)
When to use
- User pastes Weaknesses / Questions from ICLR, NeurIPS, ICML, EMNLP, etc.
- User asks how to respond to a specific reviewer comment.
- User wants a structured plan before writing rebuttal prose.
- User asks which tone or evidence strategy fits a harsh vs. open reviewer.
- User says "reviewer ignored my rebuttal" and wants to recalibrate.
When NOT to use
- User wants full rebuttal text copied verbatim → output guidance first; warn against blind LLM paste.
- User asks to evaluate a research idea → use
idea-evaluator. - User asks for pre-submission paper review → use
pre-submission-reviewer. - No review text provided → ask for it first.
Core procedure
Step 1: Inventory concerns
Extract every numbered Weakness and Question from the pasted review. If the user pasted only one reviewer's block, ask whether other reviewers exist.
Output a short table:
| ID | Type | One-line summary |
|---|---|---|
| R1.W1 | W | ... |
| R1.Q1 | Q | ... |
Step 2: Classify each concern
See: references/concern-classification.md
Tag each item as one of:
misunderstanding | evidence_gap | novelty | baseline_fairness | scope_claim | writing_clarity | theory | efficiency | other
Step 3: Match reviewer mindset (X)
See: references/mindset-matching-heuristics.md
Score the review on 5–7 behavior signals (openness, severity, constructiveness, specificity, skepticism, harshness, actionability). Map scores to a cluster in references/mindset-library.md.
State: matched cluster name, confidence (high / medium / low), and the specific phrases that triggered the match.
Step 4: Strategy priorities for this mindset (Y)
From the matched cluster in references/mindset-library.md, read:
- Historical up-rate: calibrate how likely a score increase is.
- Rule-stat diffs (success vs. failure %): boost tactics with positive diff.
- Cohen's d effects: highlight dimensions with d > 0.8 in success group.
- Success / failure patterns: concrete do's and don'ts from data.
- Key strategy: one-sentence summary for this cluster.
Cross-check dimensions in references/strategy-dimensions.md.
Step 5: Per-concern guidance
For each concern, emit a guidance block (not final prose):
### R?.W?: [short title]
- **Concern type**: ...
- **Mindset-aware strategy**: ...
- **Recommended structure**: Acknowledge → Response → Evidence → Revision note
- **Clarify from paper?**: yes/no: where to look (section, table, appendix)
- **New experiment needed?**: yes/no: what exactly (no fabricated numbers)
- **Tone note**: ...
- **Template**: A / B / C: see references/response-templates.md
- **Avoid**: ...
Do not fabricate experimental numbers. Mark anything unverified as
[TO VERIFY].
Step 6: Integrity gate
Before finalizing, run through references/checklist.md.
Key flags:
- Every W/Q has a guidance block.
- No fabricated numbers.
- Defensive phrases (e.g., "you misunderstood") flagged.
- Overpromise risk flagged for scope_claim and evidence_gap types.
Step 7: Output format
Deliver a ## Rebuttal Guidance section containing:
- Reviewer mindset summary: cluster name, confidence, identifying phrases.
- Global strategy priorities: 3–5 bullets from mindset data.
- Per-concern guidance: Step 5 blocks for each W/Q.
- Checklist highlights: pass/fail on key integrity items.
Reference map
| Purpose | File |
|---|---|
| Concern type → response priority | concern-classification.md |
| Reviewer signal → mindset cluster | mindset-matching-heuristics.md |
| Mindset clusters with stats & patterns | mindset-library.md |
| 10 strategy dimensions (Y) | strategy-dimensions.md |
| Response templates A / B / C | response-templates.md |
| Pre-submit integrity checklist | checklist.md |
FAQ
What does the rebuttal-guidance skill do?
Provides actionable peer-review rebuttal guidance (not full rebuttal prose) using conditional persuasion P(ZX,Y): match reviewer mindset X, recommend strategy Y, optionally ground in manuscript evidence and similar successful historical (review, rebuttal) pairs. Use when the user pastes reviewer comments, asks for rebuttal help, rebuttal guidance, response to reviewers, how to reply to R1.W2, OpenReview rebuttal, or the reviewer ignored the rebuttal.
How do I install rebuttal-guidance?
Run npx skills add https://github.com/hkustdial/supervisor-skills --skill rebuttal-guidance in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/rebuttal-guidance/ (personal) or .claude/skills/rebuttal-guidance/ (project).
Which AI agents can use rebuttal-guidance?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is rebuttal-guidance safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains rebuttal-guidance?
It is published by hkustdial in hkustdial/supervisor-skills. It has 326 recorded installs.
More from hkustdial/supervisor-skills
figure-designer
hkustdial/supervisor-skills
Advises on the design of the three core figures in a technical paper: the Motivated Example (Figure 1), the Solution Overview (Methodology), and the Experimental Results figures. Recommends the right design paradigm, layout, labelling, and tool for each figure type, then runs a quality-control audit. Use when the user asks to 'design a figure', 'draw Figure 1', 'plot experiment results', 'choose the right chart type', 'which figure tool to use', or 'figure looks unprofessional'.
idea-evaluator
hkustdial/supervisor-skills
Evaluates a preliminary research idea against a five-dimension framework (Higher, Faster, Stronger, Cheaper, Broader) plus idea-lifecycle and student-capability matching, paradigm-shift probing, and a fatal-flaws audit. Returns a reviewer-style verdict; non-STEM ideas route to substitute frameworks. Use when the user has a draft research idea and asks whether it is worth pursuing, asks to 'evaluate this idea', 'score this idea', 'assess feasibility', 'novelty check', 'is this a good research direction', or before committing to a paper scope.
pre-submission-reviewer
hkustdial/supervisor-skills
Runs a pre-submission review of a technical paper across five dimensions: macro logic, writing details, English grammar, LaTeX formatting, and figure quality. Uses a reviewer-style severity taxonomy (CRITICAL / MAJOR / MINOR) and flags banned AI-tone vocabulary and em-dash misuse. Use when the user asks to 'review this paper', 'audit before submission', 'check the draft', 'find issues', 'proofread', or within one week of a submission deadline.
intro-drafter
hkustdial/supervisor-skills
Drafts the Introduction prose for a technical paper, guided internally by a six-paragraph flowchart: background and running example, existing limitations, problem essence and goal, key challenges, solution overview, contributions. Positions the paper as Technique or New Problem/Setting, aligns contributions with challenges, and weaves verified citations. Outputs flowing prose by default, or an outline on request. Use when the user asks to draft, outline, or restructure an Introduction.
tech-paper-template
hkustdial/supervisor-skills
Structures a technical paper's full logical skeleton using a thinking-template table (research background, limitations, key idea or goal, challenges, methodology modules, contributions), positions the paper as Technique or New Problem/Setting, and runs a four-point self-consistency check. Use when the user is brainstorming a paper, discussing progress with an advisor, or planning the paper before drafting. Also use for 'paper skeleton', 'paper logic chain', 'thinking template', 'paper-structure planning'.
vibe-research-workflow
hkustdial/supervisor-skills
Guides AI-assisted research across three sub-flows, Vibe Coding, Vibe Figure, and Vibe Writing, with behavioural rules that keep the user in charge of academic judgment while delegating mechanical work to AI. Recommends the right tool (Cursor, Claude Code, Codex, Figma, Gemini) for the current stage. Use when the user asks 'how to use AI for research', 'Vibe Coding tips', 'AI-assisted writing workflow', 'which AI tool for this', or starts an AI-assisted work session.
Popular code quality & workflow skills
grill-me
mattpocock/skills
A relentless interview to sharpen a plan or design.
improve-codebase-architecture
mattpocock/skills
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
handoff
mattpocock/skills
Compact the current conversation into a handoff document for another agent to pick up.
domain-modeling
mattpocock/skills
Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.
codebase-design
mattpocock/skills
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.
diagnosing-bugs
mattpocock/skills
Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.