Writing Style Huasheng
v1.0.0花生的中文科技写作风格。叙事驱动、数据密集、口语化但信息密度高。 擅长用故事手法解读技术突破、人物传记和实操教程。
⭐ 0· 44·0 current·0 all-time
by@shing19
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description, and the SKILL.md all describe a writing style guide. There are no unrelated environment variables, binaries, installs, or config paths requested — everything the skill requires (nothing) is proportional to a style/template skill.
Instruction Scope
SKILL.md is a detailed style-and-template guide for generation and self-check. It does recommend preferring '一手数据' (API queries, proprietary tools, exclusive interviews) and requires '每个数字都有出处' (each figure must have a source). Those are editorial guidelines rather than explicit instructions to access specific external services, but they could lead an agent to fetch external data if the agent has browsing/API capabilities. The document itself does not instruct reading arbitrary system files, environment vars, or posting to unexpected endpoints.
Install Mechanism
No install spec and no code files — lowest-risk instruction-only skill. Nothing is downloaded or written to disk by the skill package itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. There are no disproportionate credential requests relative to the editorial writing purpose.
Persistence & Privilege
always is false (default); the skill does not request forced or permanent inclusion. disable-model-invocation is default false (agent may invoke autonomously), which is typical for skills. No instructions to modify other skills or system-wide config are present.
Scan Findings in Context
[no-findings] expected: The regex scanner had nothing to analyze because this is an instruction-only skill with no code files. That absence of findings is expected and not evidence of risk.
Assessment
This skill is a style/template guide and appears internally consistent and low risk. Things to consider before installing or allowing autonomous use: (1) If your agent has web-browsing, API, or data-access capabilities, following the guide's 'use one-hand data' and 'cite every number' rules may cause the agent to query external services or your connected accounts — ensure you are comfortable with that and that appropriate permissions/accounts are in place. (2) The guide explicitly forbids exposing real agent names, workspace paths, or private IDs in examples — keep an eye on outputs to avoid accidental leakage of internal identifiers. (3) As with any writing assistant, verify factual claims and sources the agent provides (especially numeric data) before publishing. If you want minimal risk, use this skill only when the agent's external-access capabilities are restricted or when you manually review any external lookups it performs.Like a lobster shell, security has layers — review code before you run it.
latestvk977c8fqnr4q48g8kvp02ke1c983ymwb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.