coauthor-review
Skill17 starsChanged 3 months agoMITView on GitHub
Simulated coauthor read of the full draft — biggest hole, biggest distraction, biggest opportunity, one structural suggestion, a few line edits — as a skeptical methodologist, big-picture theorist, or junior reader. Use on a complete draft before the mock referee, or when the user wants tough feedback on the paper.
In this repository38 other files it gives its agents.
- abstract-shotgunSkill
- analyzeSkill
- archiveSkill
- carefulSkill
- codebookSkill
- cover-letterSkill
- data-acquireSkill
- data-cleanSkill
- design-researchSkill
- draft-sectionSkill
- freezeSkill
- guardSkill
- hypothesis-designSkill
- idea-shotgunSkill
- identification-reviewSkill
- journal-fitSkill
- learnSkill
- lit-mapSkill
- llm-checklistSkill
- mstack-initSkill
- mstack-upgradeSkill
- paper-statusSkill
- pdf-ingestSkill
- power-analysisSkill
- preregisterSkill
- r-and-rSkill
- referee-mockSkill
- research-questionSkill
- results-auditSkill
- retroSkill
- robustnessSkill
- scope-challengeSkill
- submission-packSkill
- survey-buildSkill
- theory-buildSkill
- title-shotgunSkill
- unfreezeSkill
- vizSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Stage: write · Voice: coauthor
≈ “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
/mstack:coauthor-review
Stage: write · Voice: coauthor
On a complete draft, one round before /mstack:referee-mock. The coauthor catches structural and clarity problems a referee glosses past and the author is too close to see.
$ARGUMENTS picks the persona (default skeptical-methodologist; unrecognized → default):
| Persona | Presses on |
skeptical-methodologist | Identification, sample, SEs, robustness, overclaiming in results and discussion, line by line |
big-picture-theorist | The mechanism, scope conditions, engagement with canonical work, contribution clarity |
junior-reader | A graduate student new to the area: where the reader gets lost, jargon, unexplained acronyms, buried claims, roadmap clarity |
Procedure
- Load
paper/main.texand everypaper/sections/*.tex, plus prior coauthor reviews in.mstack/referee-cache/so you can say what changed since. - Read end-to-end first, for impressions, not corrections. Then read the referee complaints table in
${CLAUDE_PLUGIN_ROOT}/references/writing-conventions.mdand, for the skeptical methodologist, the design section of${CLAUDE_PLUGIN_ROOT}/references/identification-threats.md. - Three structural notes, one paragraph each, quoting the specific section:
- The biggest hole — where the argument is weakest.
- The biggest distraction — what to cut; often the part the author is proudest of.
- The biggest opportunity — a missing argument, framing, or analysis that would land the paper better. Specific.
- One structural suggestion: a concrete reordering, restructuring, or refocusing. The one that changes the most, not ten.
- Three to five line edits: quote the line, give the alternative, five words on why.
- Save to
.mstack/referee-cache/coauthor-<persona>-<YYYY-MM-DD>.md.
Outputs
.mstack/referee-cache/coauthor-<persona>-<date>.md.- Summary block: hole, distraction, opportunity, the one structural suggestion.
Anti-patterns
- Compliments. A coauthor who says the draft is great is not coauthoring.
- Marginalia as review. Twenty line edits and no structural notes is a copy-edit.
- Breaking persona. The junior reader does not propose identification strategies; the theorist does not fix typos.
Next
Revise (/mstack:draft-section for any broken section), then /mstack:referee-mock with a different persona before submission.
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.