Index of every page on this site: llms.txt
All skills
tristanmanchester avatar

/nature-article-writer

@3323bc9
Run it yourself

Draft, revise, or critique a primary-research manuscript for a specifically chosen Nature or Nature Portfolio journal, or deliberately develop that broad-reader scientific style. Ground claims in the supplied results and calibrate journal requirements separately. Do not impose this journal style on generic documentation, reviews, press releases, or unrelated writing.

Use this Skill: https://skilld.dev/gh/tristanmanchester/agent-skills/nature-article-writer

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesintegrity-and-compliance.md

≈631 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Integrity and submission checks

Preserve observations, raw data, numerical results, uncertainty, and analysis provenance. Keep each claim connected to the figure, method, or reference that supports it. Missing details remain missing until an author/source supplies them; never fill them with plausible approvals, replicate counts, test names, or DOIs.

AI tools and figures

Human authors remain accountable. Check the target journal's current AI policy for the specific activity, including writing, copy editing, code generation, analysis, translation, and image generation. Record tool use sufficiently to make the required disclosure accurate. Do not assume a blanket copy-editing exemption from an old policy snapshot or claim to have checked inaccessible policy text.

Do not fabricate or alter scientific observations with generative imagery. A computational plot made from traceable data is different from a generated image purporting to show an experiment; assess the actual workflow and current journal rules. Preserve original files and transformations. Disclosure never makes false evidence acceptable, and an automated style score is not a test of integrity.

Reproducibility and end matter

Verify methods, sample units and independent replication, exclusions, statistics, uncertainty definitions, software/configuration versions, and figure legends against actual records. Explain what error bars and sample counts represent. Resolve data/code repositories, versions/accessions, restrictions, and access conditions; do not invent links or claim data are public before checking.

Review the applicable reporting summary, ethics/consent, funding, author contributions, competing interests, correspondence, and availability sections. An absent heading may be a formatting issue or an inapplicable requirement; an existing heading does not establish that its declaration is true.

Editorial communication

Answer reviewer points directly and state the precise completed change/location. For work not done, describe the proposal or reason for declining; do not promise experiments, deposits, or analyses on an author's behalf. Keep a cover letter's claims consistent with the manuscript and disclosed limitations.

The local opening-length checker does not validate any of these scientific or policy obligations. Maintain an evidence-based outstanding-item list rather than turning a small static check into a submission-ready verdict.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning1mo3 checks · Risk MEDIUM
  • Gen Agent Trust Hub1mo

    The skill helps users draft scientific manuscripts for Nature Portfolio journals. It contains deceptive metadata claiming the author is OpenAI, which misrepresents its source. It also processes external drafts and notes without isolation boundaries, creating an indirect prompt injection surface, and directs the agent to execute shell commands to run provided Python diagnostic scripts.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

Signed by skilld at 3323bc9. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Steadyupdated last month
metadata
{
  "version": "3.0.0",
  "reviewed": "2026-09-13"
}
Other metadata
compatibility
File read/write access for drafts; Python 3.10+ for optional local checks. Current journal requirements need their official guide or a supplied dated copy. The checker accepts UTF-8 text/ATX Markdown, not PDF, Word, or LaTeX directly.

README badge

README badge for tristanmanchester/agent-skills/nature-article-writer