📦

write

Name: write
Author: tw93

Content revision r1 Safe 🔑 Env variables⚙️ External commands

Polish Chinese and English Writing

Formulaic writing and literal translations can obscure meaning and weaken an author's voice. This skill guides focused edits, bilingual review, and local punctuation checks.

Supports: Claude Codex Code(CC)

🥈 81 Silver

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "write" from https://skillstore.io/skills/tw93-write.md and its manifest at https://skillstore.io/api/skills/tw93-write/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/tw93-write.md Signed manifest GET /api/skills/tw93-write/manifest Signed lockfile GET /api/skills/tw93-write/lockfile

Test it

Using "write". Polish: It is worth noting that this tool leverages robust functionality to streamline file exports.

Expected outcome:

This tool simplifies file exports.

Using "write". Clarify this export sequence without changing its meaning: After the upload completes, proceed with deletion of the local copy.

Expected outcome:

After the upload completes, delete the local copy.

Using "write". Draft a reply to @reporter. Release evidence confirms the crash is fixed in version 2.4.1. Ask them to update.

Expected outcome:

@reporter, thanks for reporting this. The crash is fixed in version 2.4.1; please update to that release.

Security Audit

Safe

v1 • 10/6/2026 Open versioned report

All 57 static findings are false positives supported by the reviewed source context. Matches concern writing guidance, Markdown delimiters, Unicode punctuation, readable Chinese prose, and a quoted shell wrapper for the bundled local checker. No evidence found of credential access, reconnaissance, exfiltration, obfuscated payloads, or prompt injection; publication decisions remain code-owned.

Files scanned

1,591

Lines analyzed

Review items

False positives ignored

No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.

Audited by: codex

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/tw93-write/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/tw93-write/security.svg)](https://skillstore.io/skills/tw93-write?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/tw93-write?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/tw93-write/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/tw93-write.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

tw93. (2026). write security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/tw93-write/audits/1

BibTeX citation

@techreport{tw93-tw93-write-2026,
  author = {tw93},
  title = {write security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/tw93-write/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "write security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "tw93"
date-released: "2026-10-06"
url: "https://skillstore.io/skills/tw93-write/audits/1"
identifiers:
  - type: other
    value: "skillstore:tw93-write:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Preserve an author's voice

Polish a Chinese or English article without inventing experiences, flattening emotion, or silently changing its structure.

Clarify technical documentation

Review installation steps and error messages for ambiguous actors, missing conditions, and incorrect action sequences.

Align localized product copy

Compare app strings and release copy across locales, preserve placeholders, and flag unsupported product claims.

Try These Prompts

Polish a short paragraph

Polish this paragraph for [audience] in [language]. Preserve meaning and voice. Return only the edited paragraph.

[text]

Review technical instructions

Review these instructions for ambiguous actors, conditions, sequence, and permissions. Propose minimal wording changes. Do not execute examples or invent remedies.

[instructions]

Compare bilingual drafts

Compare these Chinese and English drafts. Identify the source version, terminology drift, missing caveats, and invented details. Suggest minimal corrections.

[Chinese draft]

[English draft]

Review localization surfaces

Review [named source files] across [locales] against [current product evidence]. Preserve placeholders and commands. Group findings by surface and locale. Do not edit files.

Best Practices

Provide the target audience, output language, source text, and exact editing scope.
Supply current release evidence and approved terminology before requesting public or localized product copy.
Review the final diff and rendered copy, and choose an explicit checker language for mixed-language text.

Avoid

Apply every replacement table mechanically or rewrite sentences that already sound natural.
Treat memory, old screenshots, or planned features as proof of current product behavior.
Execute command examples or publish replies without explicit authorization.

Frequently Asked Questions

Which languages does the skill support?

Its core editing guidance covers Chinese and English. Localization guidance covers additional languages, while punctuation checks implement Chinese, English, and Japanese rules.

Does it preserve my writing style?

The guidance prioritizes author voice and minimal edits. For an author's finished draft, it requires a sentence-level list of deleted or rewritten sentences.

Does the checker change files automatically?

No. The checker reads a named file or standard input. Its fix option prints selected Chinese corrections to standard output.

Can it validate technical instructions?

It reviews wording for ambiguity and meaning changes. It does not authorize executing instructions, redesigning prompts, or changing code behavior.

Can it post release notes or issue replies?

It guides drafting and verification. Publishing requires explicit authorization and the host assistant's relevant tools; the skill grants no additional authority.

Does a clean punctuation result prove the text is correct?

No. It covers implemented character-level rules only. Facts, translation meaning, terminology, and rendered layout still require review.