engineering-response
Skill3 starsChanged 40 days agoMITView on GitHub
Draft, audit, or revise English response letters and revision plans for engineering papers from advisor, senior-author, reviewer, or editor comments. Use when the user provides comments, decision letters, rebuttal notes, revision drafts, or asks how to respond, what to revise, how to classify comments, how to avoid over-editing, or how to map each comment to manuscript changes.
In this repository7 other files it gives its agents.
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 9 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill to turn comments into traceable revision tasks and professional English responses.
- Distinctive: line 17
Do not invent experiments, line numbers, figures, citations, analyses, or manuscript changes.
- Distinctive: line 19
Prefer concise, evidence-linked responses over defensive explanations.
- Distinctive: line 20
When a comment reveals misunderstanding, first check whether the manuscript caused it.
- Distinctive: line 28
Do not force an incomparable system into the results table to satisfy a comparison request.
- Distinctive: line 35
Use this skill for comment triage, revision planning, rebuttal drafts, and response letters.
- Very specific: line 37
Use engineering-writing for manuscript section drafting after a comment has been turned into a concrete edit.
- Very specific: line 39
Use engineering-polishing for prose-only improvements.
- Very specific: line 40
Use engineering-figure-table for caption, figure, or table revisions.
- Very specific: line 41
Use engineering-validation before claiming a response package is complete or before using final line numbers.
- Distinctive: line 78
If the response claims completed changes, run the diff-verification gate.
- Distinctive: line 80
If the change is planned but not done, draft a plan or author-input note, not a final completed-change response.
- Distinctive: line 82
Run completeness and factuality checks before calling the package ready.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Engineering Response
Use this skill to turn comments into traceable revision tasks and professional English responses.
Core Stance
- Preserve each comment before responding.
- Answer every concern or mark it as unresolved.
- Map responses to manuscript evidence, a revision location, or explicit author input needed.
- Do not invent experiments, line numbers, figures, citations, analyses, or manuscript changes.
- Prefer concise, evidence-linked responses over defensive explanations.
- When a comment reveals misunderstanding, first check whether the manuscript caused it.
- Separate misreadings from real gaps. A misreading (a fallback mistaken for the main mechanism, a reused component read as added hardware) is fixed in the manuscript framing: an operating-envelope paragraph, a clearer split between primary and fallback mechanisms, a corrected word. A real gap (an unfair baseline, an unmeasured component, missing cost) needs a control or measurement when feasible, or an explicit limitation when not.
- Do not force an incomparable system into the results table to satisfy a comparison request. Discuss it qualitatively, state the differences in task, data, and hardware, and keep the controlled comparison as the main evidence. A weak baseline can be reframed as a reference configuration.
Boundaries
- Use this skill for comment triage, revision planning, rebuttal drafts, and response letters.
- Use
engineering-writingfor manuscript section drafting after a comment has been turned into a concrete edit. - Use
engineering-polishingfor prose-only improvements. - Use
engineering-figure-tablefor caption, figure, or table revisions. - Use
engineering-validationbefore claiming a response package is complete or before using final line numbers.
When to Open Extra Files
| File | Open when |
| references/comment-routing.md | Classifying advisor/reviewer comments and deciding which section to edit first |
| references/revision-tracker.md | Turning comments into concrete edit tasks with acceptance evidence |
| references/comment-resolution-worksheet.md | Building a full comment-to-action worksheet before editing |
| references/response-letter.md | Drafting point-by-point English responses |
| references/comment-examples.md | Handling common comments about experiments, methods, claims, captions, related work, conclusions, or abstract |
| references/tone-and-risk.md | Handling disagreement, impossible requests, missing experiments, or high-risk claims |
| references/contradictory-reviewer-strategy.md | Reviewers ask for conflicting detail, compression, experiments, or framing |
| references/partial-compliance-response.md | Authors can only partially satisfy a reviewer request |
| references/rebuttal-vs-revision-mode.md | Deciding whether the output is rebuttal, revision response, or camera-ready note |
| references/diff-verification.md | A response says a manuscript change, line number, experiment, figure, or table has already been added |
| references/examples.md | Needing concrete response tracker and reply examples |
| references/failure-modes.md | Handling false completed-change claims or impossible reviewer requests |
| ../_shared/evidence-boundary.md | A response may claim unsupported experiments, citations, or edits |
| ../_shared/citation-boundary.md | A response mentions added or corrected references |
| ../_shared/citation-verification-workflow.md | A response relies on externally discovered references that have not passed a verification gate |
| ../_shared/claim-strength.md | A response or manuscript change needs claim downgrading |
| ../_shared/list-to-argument.md | A comment batch arrives as an unordered list |
| ../_shared/output-mode.md | The user asks for response text only |
| ../_shared/terminology-ledger.md | A comment asks for renaming methods, metrics, or categories |
Workflow
- Split editor/advisor/reviewer comments into stable IDs.
- Classify each comment by type, severity, section, and missing input.
- Identify the real complaint, not just the surface wording.
- Choose
revise,defer with reason, orno change with reason. - Define prohibited over-edit.
- Define acceptance evidence and minimum verification.
- Decide whether the mode is rebuttal, revision response, camera-ready note, or internal revision plan.
- If the response claims completed changes, run the diff-verification gate.
- Draft the response only after the change or placeholder is clear. If the change is planned but not done, draft a plan or author-input note, not a final completed-change response.
- Run completeness and factuality checks before calling the package ready.
Default Output
Response strategy summary
- Package status:
- Main risks:
- Response mode:
- Validation needed before final response:
Comment-response tracker
| ID | Original comment | Type | Severity | Real complaint | First target | Second target | Action | Evidence/change needed | Prohibited over-edit | Acceptance evidence | Verification | Line-number status | Status |
Allowed status values: `Done with evidence`, `Planned`,
`Needs author input`, `Defer with reason`, `No change with reason`, and
`Not supported`.
Draft response
[English point-by-point response]
Unverified items
- line numbers:
- experiments:
- citations:
- figures/tables:
Revision checklist
- ...Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

