← 返回 Skills 市场
110
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install writing-credibility-auditor
功能描述
Audit any piece of writing for missing citations, unsupported claims, logical fallacies, weasel words, and misleading statistics — then produce a structured...
安全使用建议
This skill is internally consistent and low-risk because it only contains written instructions for analyzing text and requests no installs, files, or credentials. Before installing, note these practical limitations: the skill does not perform live fact-checking or verify external sources (it audits reasoning and language only), so outputs that appear to cite studies or assert factual corrections should be double-checked against primary sources. Also be aware that purely model-driven detection can produce false positives/negatives and may sometimes overconfidently label passages — treat the Credibility Report as guidance, not authoritative proof. If you need authoritative fact verification, prefer a skill that explicitly uses vetted web APIs or human expert review. Finally, if you later see this skill ask for environment variables, external URLs, or to run system commands, stop and re-evaluate — those would be unexpected for this skill and would raise concern.
能力评估
Purpose & Capability
The name/description (credibility auditing) match the SKILL.md and README: the skill performs language- and reasoning-based scans (fallacies, unsupported claims, weasel words, misleading stats). It requests no unrelated credentials, binaries, or configuration.
Instruction Scope
SKILL.md contains explicit, bounded instructions for analyzing user-supplied text and specifies it does not perform live web searches or other external data collection. It does not instruct the agent to read system files, environment variables, or transmit data to external endpoints.
Install Mechanism
No install spec and no code files — the skill is instruction-only, so nothing is downloaded or written to disk. This is low-risk and proportionate for a reasoning-only auditor.
Credentials
The skill declares no required env vars, credentials, or config paths. That is appropriate for a pure-reasoning text audit and there are no hidden requests in SKILL.md or README.
Persistence & Privilege
always:false (default) and model invocation is allowed (disable-model-invocation:false). Autonomous invocation is platform-default and is not in itself a security concern here — just be aware the agent can call the skill when user intent matches the triggers.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install writing-credibility-auditor - 安装完成后,直接呼叫该 Skill 的名称或使用
/writing-credibility-auditor触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
**Initial Release – writing-credibility-auditor 1.0.0**
- Launches a writing audit tool to check for missing citations, unsupported claims, logical fallacies, weasel words, and misleading statistics.
- Produces a structured credibility report highlighting issues with flagged excerpts, fallacy names, severity ratings, and suggested fixes.
- Designed for users needing critical review of articles, essays, reports, arguments, or research summaries.
- Supports detailed user guidance, including when to trigger or not trigger the skill.
- Self-contained analysis without relying on external APIs or live web search.
元数据
常见问题
Writing Credibility Auditor 是什么?
Audit any piece of writing for missing citations, unsupported claims, logical fallacies, weasel words, and misleading statistics — then produce a structured... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 110 次。
如何安装 Writing Credibility Auditor?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install writing-credibility-auditor」即可一键安装,无需额外配置。
Writing Credibility Auditor 是免费的吗?
是的,Writing Credibility Auditor 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Writing Credibility Auditor 支持哪些平台?
Writing Credibility Auditor 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Writing Credibility Auditor?
由 arbazex(@arbazex)开发并维护,当前版本 v1.0.0。
推荐 Skills