← 返回 Skills 市场
178
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install translate-agent
功能描述
Stateless translation and summarization machine. Always respond with raw JSON only — no markdown, no explanation, no preamble, no code fences. Trigger this s...
安全使用建议
This is an instruction-only translation/summarization skill and appears coherent and low-risk: it asks for nothing beyond the agent following its JSON schemas. Before installing, consider that (1) it enforces strict JSON-only outputs which may break workflows expecting natural-language replies, (2) it mandates being triggered for inputs that look like translation requests — test how it interacts with other skills or handlers in your agent, and (3) because it is instruction-only, all translation is performed by whatever model or system the agent already uses (so avoid sending highly sensitive content through it if your agent's model logging or telemetry is a concern). Test with non-sensitive examples (including edge cases like very long texts and unknown BCP-47 codes) to confirm behavior matches your expectations.
功能分析
Type: OpenClaw Skill
Name: translate-agent
Version: 1.0.0
The skill bundle defines a structured translation and summarization utility. The instructions in SKILL.md focus entirely on processing text inputs and returning formatted JSON responses, with no evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
Name and description match the instructions: the SKILL.md describes only translation/summarization behaviors and output schemas. The skill requests no binaries, env vars, or config paths that would be unrelated to translation.
Instruction Scope
All runtime instructions stay within translation/summarization scope (detect input, dispatch by action, validate, produce JSON). The skill does not instruct reading files, environment variables, or contacting external endpoints. It enforces strict JSON-only outputs, which is a behavioral constraint but not a security risk by itself.
Install Mechanism
No install spec and no code files are present (instruction-only). Nothing is written to disk and no external downloads are requested.
Credentials
The skill requires no environment variables, credentials, or config paths. There are no disproportionate secret/credential requests relative to the described functionality.
Persistence & Privilege
always is false and the skill does not request persistent/system-level settings. Autonomous invocation is allowed (platform default) but this skill does not grant elevated privileges or modify other skills.
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 translate-agent - 安装完成后，直接呼叫该 Skill 的名称或使用
/translate-agent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 of TranslateAgent skill.
- Provides stateless translation and summarization with deterministic, raw JSON output only.
- Supports actions: "translate", "summarize", "translate_and_summarize", and "heartbeat".
- Automatically detects input type and language, defaulting to Vietnamese for translation.
- Returns structured JSON responses for each action; robust error handling included.
- Strict meta rules enforce output consistency and schema integrity.
元数据
常见问题
Translate Agent 是什么？
Stateless translation and summarization machine. Always respond with raw JSON only — no markdown, no explanation, no preamble, no code fences. Trigger this s... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 178 次。
如何安装 Translate Agent？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install translate-agent」即可一键安装，无需额外配置。
Translate Agent 是免费的吗？
是的，Translate Agent 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Translate Agent 支持哪些平台？
Translate Agent 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 Translate Agent？
由 AnhducNA（@anhducna）开发并维护，当前版本 v1.0.0。
推荐 Skills