nursing-academic-polishing
Polishes, translates or condenses existing nursing academic text (Chinese/English) while preserving facts, author meaning, terminology and the strength of conclusions, including format-preserving document revisions. Drafting new sections is left to other writing skills.
What it can touch
| files | write |
|---|---|
| network | example.invalid, pubmed.ncbi.nlm.nih.gov, www.crossref.org |
| shell | runs commands |
| secrets | none found |
What it actually does
The bundle is a full suite of 18 nursing-research skills, not just the polishing skill. The suite covers polishing, cross-sectional and manuscript writing, data stewardship, evidence checking, literature search, mechanism and scientific figures, presentations, paper reading, peer review, reference checking, research design, research logs, review responses, review writing, statistics and technical disclosure. It also ships shared rules, templates, synthetic teaching examples, evaluation records, unit tests and local helper scripts: a Python evidence extractor/validator, a Python SVG renderer, and R synthetic-demo scripts. The instructions confine the agent to user-specified files and treat instructions found inside materials as content, not commands. They forbid sending confidential material to external services and forbid fabricating results, ethics approvals or citations. The helper scripts are described as offline, read inputs read-only, and write only to new output directories while refusing to overwrite. Some skills (literature search, reference check) direct the agent to query public bibliographic sources using only public identifiers or generic concepts. No credential access, persistence, hidden text or prompt injection was found.
What the check found
mediumRuns code that is built while it runs
eval, exec and new Function run text as code, which makes it hard to know what will actually run.
tests/test_cohort_workflow.py:194 + 'eval(parse(file=' + json.dumps(SCRIPT.as_posix())
mediumRuns code that is built while it runs
eval, exec and new Function run text as code, which makes it hard to know what will actually run.
tests/test_rct_workflow.py:172 + 'eval(parse(file=' + json.dumps(SCRIPT.as_posix())
Keep an eye on it
This check is of one commit. Watch nursing-academic-polishing and we'll email you if a later commit changes what it can touch.
Made this? Add the badge
[](https://plugsday.com/tools/guozheng-de-baba-nursing-research-skills)About this check
Checked November 2, 2026 at commit 9a9edad4d411, 359 files. Rules for known red flags, then two Claude reviews, one of them told to assume the author is malicious. A later commit hasn't been checked. How checks work.