create skill
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill coherently helps create OpenClaw skills, with the main thing to notice being that it may persist new skill files and registry entries.
This appears appropriate for an instruction-only skill creator. Before using it, review any generated skill files, optional scripts, and registry updates, because those changes can persist and affect future OpenClaw behavior.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A newly created skill or registry entry could influence future agent behavior if the generated content is incorrect or too broad.
Creating files and updating the skill registry persists changes that can affect future OpenClaw behavior; this is purpose-aligned and paired with a confirmation step, so it is a note rather than a concern.
文件生成阶段 - 根据选择生成相应文件 - 更新技能注册表
Review the proposed files and registry changes before confirming, keep changes scoped to the intended skills/<技能名> directory, and retain a way to revert them.
If the user chooses a script-based skill, generated code could run local commands or access local files depending on its contents.
The skill may generate JS or shell files and provides a user-run test command; this is expected for creating executable skills but should be reviewed before execution.
可选：`<技能名>.js`（如果需要Node.js实现）或 `<技能名>.sh`（如果需要Shell实现）... 如有脚本，运行测试：`cd skills/<技能名> && node <技能名>.js`
Prefer documentation-only skills when sufficient; if scripts are generated, inspect them and ensure dependencies and required binaries are clearly declared before running.