← 返回 Skills 市场
samadhifire

zijiyong

作者 SamadhiFire · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
106
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install zijiyong
功能描述
Use when the user mentions wos, WOS, WoS, or Web of Science and wants topic-based literature search, Shenzhen University library login, paper screening, abst...
安全使用建议
This skill appears to do what it says (search WoS via the Shenzhen University route and write to Feishu via local lark-cli), but there are a few things to check before installing/using it: - Confirm you trust the skill's source: there is no homepage or repository link and the owner is unknown. Prefer skills with clear provenance. - The runtime needs local tools (lark-cli and possibly a playwright-cli). The skill metadata lists no required binaries — verify those CLIs are installed and configured locally before use. The skill will run lark-cli commands that can read or modify your Feishu Base via your local CLI session. - The skill asks for Shenzhen University (SZU) login credentials and will pause for 2FA; do not share persistent passwords. The SKILL includes a hardcoded default SZU username (2410032057) — ask the author why that value is present and remove it if it is unrelated to you. - Because the skill runs local commands and interacts with third-party sites (SZU auth, WoS, Feishu via lark-cli), only proceed if you are comfortable providing the necessary runtime credentials and have verified the local tooling is trustworthy and properly scoped. If you want to proceed: ask the skill author to (1) declare required binaries (lark-cli, playwright-cli if used), (2) remove or explain any hardcoded personal identifiers, and (3) provide a source or homepage so you can verify provenance. If these are not available, treat the skill as higher-risk and prefer manual execution of the described workflow instead.
功能分析
Type: OpenClaw Skill Name: zijiyong Version: 1.0.1 The skill bundle facilitates a legitimate academic workflow for searching Web of Science (WoS) via the Shenzhen University (SZU) library and exporting results to Feishu Base using local CLI tools. It demonstrates good security practices by explicitly instructing the agent not to store passwords, to wait for user intervention during multi-factor authentication (MFA), and to confirm target locations before writing data. While it contains specific local environment details like a default SZU username and local file paths (e.g., in SKILL.md and playbook.md), these appear to be for the user's personal automation convenience rather than malicious intent.
能力评估
Purpose & Capability
The skill is clearly intended to search Web of Science via the Shenzhen University library and write results into Feishu Base using a local lark-cli; that purpose is coherent with the description. However, the runtime instructions depend on local tooling (lark-cli, optionally playwright-cli/playwright-cli skill) and access to the user's SZU login, yet the skill metadata declares no required binaries or credentials. The absence of declared required binaries (lark-cli, browser automation) is an incoherence that could surprise users or lead the agent to attempt commands that will fail or prompt for credentials unexpectedly. Also the SKILL contains a hardcoded default SZU username (2410032057), which appears tailored to a specific user/workflow and is unexpected in a generic published skill.
Instruction Scope
SKILL.md and the playbook tightly specify asking preflight questions, building WoS queries, pausing for second-factor verification, and performing local lark-cli commands to create/update Feishu Base records. These instructions stay within the advertised purpose (WoS -> Feishu), but they intentionally require accessing the user's local CLI, local file paths, and the user's Shenzhen University credentials (runtime-only). That behavior is reasonable for the task, but because the skill will direct the agent to run local commands and potentially ask for credentials, the user should be aware and confirm those steps. The hardcoded SZU username is an unexpected detail and could be a privacy/leak concern.
Install Mechanism
This is an instruction-only skill with no install spec, so nothing will be written to disk by the skill package itself (low install risk). However, the runtime steps assume the presence and correct configuration of external CLI tools (lark-cli and possibly playwright-cli). Because these binaries are not declared in the skill metadata, the agent may fail or prompt the user to install them. No external download URLs or install scripts are present in the repo.
Credentials
The skill declares no required environment variables or credentials, and it does not request unrelated service keys. It does instruct the agent to ask the user for SZU login credentials at runtime (and to treat them as transient), which is proportionate to the stated purpose. Still, the presence of a specific default username (2410032057) is unusual and should be confirmed with the skill author or removed. There is no evidence the skill would exfiltrate other environment variables, but the agent will interact with local CLIs that hold the user's Feishu authorization state, so the user should confirm the local CLIs are trusted.
Persistence & Privilege
The skill does not request permanent presence (always: false) and does not declare modifications to other skills or global agent settings. It explicitly instructs not to persist passwords or verification codes and to treat credentials as runtime-only. Autonomous invocation is allowed by default (disable-model-invocation: false), which is normal; this is not combined with other high-risk flags.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install zijiyong
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /zijiyong 触发
  4. 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
v1.0.1
wos-feishu-literature 1.0.1 - Introduced a new skill for topic-driven literature search and Feishu Base writeback, triggered by "WOS", "Web of Science", etc. - Enforces a structured workflow: asks preflight questions, prefers Shenzhen University library route, and defaults to SSCI database. - Requires user confirmation for Feishu table details and handles multi-step authentication pauses. - Specifies detailed rules for screening, deduplication, field extraction, and JCR Quartile lookups. - Prioritizes local lark-cli for Feishu integration; supports both Windows and PowerShell edge cases.
元数据
Slug zijiyong
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

zijiyong 是什么?

Use when the user mentions wos, WOS, WoS, or Web of Science and wants topic-based literature search, Shenzhen University library login, paper screening, abst... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 106 次。

如何安装 zijiyong?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install zijiyong」即可一键安装,无需额外配置。

zijiyong 是免费的吗?

是的,zijiyong 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

zijiyong 支持哪些平台?

zijiyong 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 zijiyong?

由 SamadhiFire(@samadhifire)开发并维护,当前版本 v1.0.1。

💬 留言讨论