From draft-detective
Lists all Draft Detective capabilities and asks the user which check to run. Use when the user gives a vague request like 'analyze my document' or 'run draft detective.'
How this skill is triggered — by the user, by Claude, or both
Slash command
/draft-detective:capabilitiesThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The user has asked to use Draft Detective but has **not** specified what to run. Draft Detective offers several capabilities, each implemented as its own skill. Your job here is **only to help them choose** — do not run anything yet, and do not guess on their behalf.
The user has asked to use Draft Detective but has not specified what to run. Draft Detective offers several capabilities, each implemented as its own skill. Your job here is only to help them choose — do not run anything yet, and do not guess on their behalf.
Keep your message concise and scannable — a short intro line, the lists, and the question.
These review the document and flag issues.
| Assessment | What it checks | Skill |
|---|---|---|
| Reference Error Checker | Are your references accurate? Uses web search to confirm each citation exists and that the author, title, publisher, and year match public sources — catching typos and hallucinated references. | reference-validation |
| Figures & Tables Check | Are all figures and tables properly titled, consistently numbered, cited in the body text, and do all body-text references resolve to a real figure or table? | figures-tables-check |
| Document Contents | Does the document include all required sections — About This, Acknowledgements, Methods, Results, Conclusion, References, and an Appendix when one is referenced? | document-contents |
| Recommendation Check | Is each recommendation supported by the document's own findings? Flags recommendations with weak, indirect, missing, or contradictory backing. | recommendation-check |
| Internal Inference Validation | Does the reasoning hold up? Flags logical leaps, unsupported conclusions, and arguments where the evidence doesn't support the claim. | inference-validation |
| Advocacy & Tone | Does the document use neutral, objective language? Flags trigger words (certainty language without evidence), advocacy language (unsupported recommendations or opinion framing), and subjective tone. (Beta.) | advocacy-tone |
| Reviewer 2 | A rigorous simulated peer review — summary, strengths, weaknesses, and prioritized next steps — plus a separate devil's-advocate rebuttal. (Beta.) | reviewer-2 |
| Literature Review | Are there relevant sources you may have missed? Searches the web for academic sources related to your document's claims — both supporting and conflicting — that aren't already cited. (Beta.) | literature-review |
| Live Reports | Have your findings been updated or contradicted by newer research? Searches the web for sources published after your document's date and produces an addendum of what to update. (Beta.) | live-reports |
| Methodological Alignment | Does your methodology match standard practice in the field? Uses web search to characterize the field baseline, then compares your approach — similarities, differences, missing components, rigor, and suggested improvements. | methodology-comparison |
| Reproducibility Check | Could someone reproduce your results from the document alone? Extracts the main results and classifies each by how reproducible it is. (Beta.) | reproducibility-check |
These operate on the document rather than flagging issues.
| Tool | What it does | Skill |
|---|---|---|
| Download References | Search the web for a reference and download its full original content (PDF/Markdown), verifying the match. Reports found / found-but-inaccessible / not-found. | reference-download |
| Extract Methodology | Extract a structured, reproducible description of the paper's methodology, and classify how reproducible it is. | methodology-extraction |
| Extract References | Find and list all bibliographic references from the document's reference/bibliography section. | reference-extraction |
npx claudepluginhub agencyenterprise/draft-detective --plugin draft-detectiveSystematic self-review checklist for academic papers covering structure, logic consistency, citations, claim auditing, figure/table quality, and writing clarity.
Audits academic or technical manuscripts with a section-level refactoring report covering argument architecture, narrative flow, citation hygiene, and submission-readiness.
Audits academic papers pre-submission using parallel agents for content, numerical consistency, references/DOIs, writing quality, figures/formatting, and replication archives.