← 返回 Skills 市场
xiaojia-skills
作者
tanchaowen84
· GitHub ↗
· v1.0.0
· MIT-0
110
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install xiaojia-skills
功能描述
Call the JustAI openapi async chat endpoints and return structured JSON results. Use when Codex needs to invoke the deployed JustAI agent for marketing plans...
安全使用建议
Before installing or using this skill:
- Expect to set JUSTAI_OPENAPI_BASE_URL and JUSTAI_OPENAPI_API_KEY in your environment; the code will exit if they're not present even though the registry metadata doesn't list them. This mismatch is likely an oversight but you should be aware.
- The scripts will send your prompts, conversation_id, and any provided project_id/skill_id to the configured base URL. Only use a base URL and API key you trust (preferably a well-audited internal or vendor endpoint).
- Inspect the bundled scripts (they are small and readable) before running them in a sensitive environment. They use urllib to POST JSON; no other hidden network calls or file exfiltration are present in the code.
- Limit the scope of the API key you provide (use least privilege) and avoid putting broad or long-lived org-level keys into this skill if possible.
- Because the source/homepage is unspecified in the registry, consider installing from the upstream repository referenced in README (verify its authenticity) or obtain the code via a verified source.
- If you need stricter controls, run the scripts in an isolated environment/container or use a proxy to monitor outbound requests to the configured base URL.
If you want, I can list the exact lines that require the env vars and the network endpoints the scripts will call.
功能分析
Type: OpenClaw Skill
Name: xiaojia-skills
Version: 1.0.0
The xiaojia-skills bundle is a legitimate integration for the JustAI OpenAPI, providing scripts (chat.py, list_projects.py, etc.) to perform asynchronous chat tasks and resource management. It correctly handles API authentication via environment variables and uses standard Python libraries for network communication without any evidence of malicious behavior, data exfiltration, or prompt injection.
能力评估
Purpose & Capability
Name/description claim: call JustAI openapi async chat endpoints. The bundled Python scripts implement exactly that (list projects/skills, submit chat, poll results). However the registry metadata listed no required environment variables while the SKILL.md and scripts require JUSTAI_OPENAPI_BASE_URL and JUSTAI_OPENAPI_API_KEY — an inconsistency in declared requirements versus actual needs.
Instruction Scope
Runtime instructions and the scripts are narrowly scoped: they only read the specified JUSTAI_OPENAPI_* env vars, send JSON payloads to the configured base URL, and print JSON results. They do not read other system files, other credentials, or perform arbitrary shell commands beyond invoking the included Python scripts. Note: using the scripts will transmit user messages and any supplied project_id/skill_id to the configured external endpoint.
Install Mechanism
There is no install spec — this is effectively an instruction-only skill bundled with Python scripts. That is low-risk from an automatic-install perspective, but the bundled scripts will be written into the skills directory and executed when invoked; review the shipped code before running.
Credentials
The scripts reasonably require JUSTAI_OPENAPI_BASE_URL and JUSTAI_OPENAPI_API_KEY (and optional JUSTAI_OPENAPI_TIMEOUT). That matches the skill's networked purpose, but the package/registry metadata listed no required env vars — a mismatch that could mislead users or automated guards about the need to provide an API key and base URL. Also, providing the API key gives the skill (and anyone with access to that key) the ability to call the remote JustAI service with your data, so key scope and trust in the endpoint matter.
Persistence & Privilege
Flags indicate the skill is not forced always-on and it doesn't request special system-wide persistence. It does not modify other skills or system settings. The skill can be invoked autonomously by the agent (default platform behavior) — not a problem in isolation but worth noting when combined with external network access.
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 xiaojia-skills - 安装完成后,直接呼叫该 Skill 的名称或使用
/xiaojia-skills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 xiaojia-skills for structured marketing and content automation via JustAI openapi async chat endpoints.
- Provides Python scripts to list projects, list skills, submit chat tasks, and poll results with clear CLI workflows.
- Supports asynchronous, two-step chat operations (submit and poll) for tasks like marketing plan generation, Xiaohongshu notes, image generation, collect-info and confirm-info flows.
- Standardizes use of environment variables and enforces use of project_id and skill_id arrays for scoping requests.
- Introduces guardrails: conversation_id management, natural-language continuation for card confirmation, and error surfacing on failed tasks.
- Includes detailed usage examples for running all supported operations from the shell.
元数据
常见问题
xiaojia-skills 是什么?
Call the JustAI openapi async chat endpoints and return structured JSON results. Use when Codex needs to invoke the deployed JustAI agent for marketing plans... 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 xiaojia-skills?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install xiaojia-skills」即可一键安装,无需额外配置。
xiaojia-skills 是免费的吗?
是的,xiaojia-skills 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
xiaojia-skills 支持哪些平台?
xiaojia-skills 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 xiaojia-skills?
由 tanchaowen84(@tanchaowen84)开发并维护,当前版本 v1.0.0。
推荐 Skills