← Back to Skills Marketplace
321
Downloads
0
Stars
1
Active Installs
7
Versions
Install in OpenClaw
/install blog-polish-zhcn
Description
Polish and translate a technical blog draft into a 1200–1400 word, 4-5 section Markdown article in Simplified Chinese (zh-CN), preserving technical terms and...
README (SKILL.md)
Blog Polish (zh-CN)
This skill polishes a technical blog draft and translates it to Simplified Chinese while preserving technical accuracy.
When to Use
Use when the user asks to polish/translate a technical blog draft to zh-CN without images. Triggers: "polish my draft", "translate blog to Chinese", "enhance latestDraft.md".
Defaults
draftPath:~/.openclaw/workspace/contentDraft/latestDraft.mdoutputDir:~/.openclaw/workspace/contentPolished/- Output filename:
${ts}_${titleSlug}.md
Workflow Summary
- Resolve paths + create timestamp (
date +"%y%m%d%H%M") - Read draft from
draftPath - Polish English: Fix grammar/spelling, improve clarity, structure into 4-5 sections, target 1000-1200 words
- Translate to zh-CN: Preserve code blocks, inline code, technical terms (
openclaw,skill,cli) - Save polished markdown to
${outputDir}/${ts}_${titleSlug}.md - Return:
{ outputPath: "/full/path/to/file.md" }
Output
The skill returns a JSON object with a single property:
outputPath: path to the generated polished article.
Example
Input: ~/.openclaw/workspace/contentDraft/latestDraft.md
Example Output
{ "outputPath": "~/.openclaw/workspace/contentPolished/2603142134-openclawSkills.md" }
Usage Guidance
This skill appears to do what it says: read a single markdown draft, send it to the platform LLM to be polished and translated, then save the result. Before installing or running it: 1) Confirm whether your environment has 'jq' installed (SKILL.md lists jq as required); if not, either install jq or update the skill/finalize step. 2) Verify the draftPath and outputDir defaults point only to locations you expect (avoid pointing draftPath at sensitive files). 3) Remember that the 'llm: true' step will transmit the draft text to whatever model/provider your OpenClaw agent is configured to use — do not use this skill on content you do not want sent to that model. 4) Consider resolving the word-count inconsistency (1200–1400 vs 1000–1200) so outputs meet your expectations.
Capability Analysis
Type: OpenClaw Skill
Name: blog-polish-zhcn
Version: 1.0.13
The skill is designed to polish and translate technical blog drafts into Simplified Chinese. The workflow in SKILL.md uses standard shell commands (cat, mkdir, printf, jq) and LLM processing to handle files within a designated workspace (~/.openclaw/workspace/). There are no indicators of data exfiltration, malicious execution, or harmful prompt injection.
Capability Assessment
Purpose & Capability
The skill's stated purpose (polish and translate a markdown draft to Simplified Chinese) matches what the workflow does: it reads a draft file, sends the text to the platform LLM step, and writes a polished markdown file. Minor inconsistencies exist in the metadata: SKILL.md lists jq as a required tool in its internal metadata, but the registry-level 'Required binaries' field was empty. Also the description says 1200–1400 words while the workflow summary mentions 1000–1200 words — clarify desired target.
Instruction Scope
Runtime instructions only read the specified draftPath and write the polished file to outputDir, which is appropriate. The workflow uses an 'llm: true' step — the draft text will be sent to the configured LLM provider for processing, so any sensitive content in the draft will be transmitted according to your platform's model provider settings. The skill does not instruct reading other files or environment variables.
Install Mechanism
There is no install spec and no code files (instruction-only), so nothing is written to disk beyond the output file the skill creates. This is low-risk. Note: SKILL.md's embedded metadata declares a dependency on 'jq' but there is no install step to fetch it; the platform should already provide jq or the finalize step will fail.
Credentials
The skill requests no environment variables or credentials. It reads a user-specified file path (defaulting to a workspace path) and writes to a user-specified output directory — these are proportionate to the task. Users should ensure the draftPath points only to the intended draft (don’t point it at sensitive system files).
Persistence & Privilege
The skill does not request persistent/always-on privileges (always: false). It uses save_state/load_state calls to store transient workflow state (timestamp, paths, titleSlug), which is normal for a skill run and limited in scope.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install blog-polish-zhcn - After installation, invoke the skill by name or use
/blog-polish-zhcn - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.13
- Increased target article length to 1200–1400 words.
- Output filename now includes a title slug for better clarity (e.g., `${ts}_${titleSlug}.md`).
- Improved workflow: extracts a slugified title and separates file writing from content generation.
- Updated metadata for schema version 1.1.
- Removed unnecessary files to streamline the skill directory.
v1.0.12
blog-polish-zhcn 1.0.12
- Initial version of the zh-CN specific blog polish skill.
- Polishes and translates a technical blog draft to Simplified Chinese (zh-CN), preserving technical terms and code blocks.
- Provides a simplified workflow focused on Chinese output (no Astro/English output).
- Adds metadata and documentation tailored for Chinese blog polishing and translation use cases.
v1.0.11
- Fixed output directory variable to expand environment variables correctly in workflows.
- Version bump to 1.0.11.
- No changes to features, workflow, or outputs.
v1.0.10
- Major update: The skill now supports polishing and translating technical blog drafts into both English and Simplified Chinese, and outputs an Astro markdown compatible version.
- Adds workflow steps for generating an Astro-compatible slug, preparing Astro frontmatter, and creating both English and Chinese outputs.
- Output now includes separate English, Simplified Chinese, and Astro markdown paths.
- New documentation files added for multilingual and Astro markdown support.
v1.0.9
- Changed timestamp format in the workflow from `%Y%m%d%H%M` to `%y%m%d%H%M` for output filenames.
- No other changes detected.
v1.0.8
- Added entrypoint metadata field for improved integration and skill discovery.
- Fixed init step to ensure proper date formatting and state saving.
- Added set_output: outputPath to polish_and_translate step for consistent output handling.
- Version bump to 1.0.8.
v1.0.7
Initial release of blog-polish-zhcn:
- Polishes and translates technical blog drafts into Simplified Chinese (zh-CN) Markdown articles (1000–1200 words, 4–5 sections).
- Preserves technical terms and code blocks.
- Supports default draft and output paths; outputs polished file path.
- Designed for technical content enhancement and translation workflows.
Metadata
Frequently Asked Questions
What is blog-polish-zhcn?
Polish and translate a technical blog draft into a 1200–1400 word, 4-5 section Markdown article in Simplified Chinese (zh-CN), preserving technical terms and... It is an AI Agent Skill for Claude Code / OpenClaw, with 321 downloads so far.
How do I install blog-polish-zhcn?
Run "/install blog-polish-zhcn" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is blog-polish-zhcn free?
Yes, blog-polish-zhcn is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does blog-polish-zhcn support?
blog-polish-zhcn is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created blog-polish-zhcn?
It is built and maintained by Jeff Yang (@j3ffyang); the current version is v1.0.13.
More Skills