← 返回 Skills 市场
Uplo Hr
作者
RooJenkins
· GitHub ↗
· v1.0.0
· MIT-0
200
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install uplo-hr
功能描述
AI-powered HR knowledge management. Search employee handbooks, org charts, company policies, benefits documentation, and onboarding materials with structured...
安全使用建议
Before installing, verify the following: (1) Confirm provenance of the package and publisher: ask for the official homepage/source repo and verify the npm package `@agentdocs1/mcp-server` (or the referenced package) is legitimate and audited. (2) Expect to provide an UPLO instance URL and a secret API_KEY — make sure the endpoint is your managed internal instance (not a public service) and apply least privilege to the token. (3) Running `npx` will download and execute remote code; if your org forbids executing external npm packages, block/install only after security review. (4) Ensure identity scoping and access control are enforced so search results don’t leak personnel records to unauthorized users. (5) Resolve the metadata inconsistency (registry lists no required env vars while skill.json requires them) — ask the publisher to correct the manifest or provide clear install/setup docs. If you cannot verify the npm package or publisher, or your security policy forbids executing third-party code, do not install.
功能分析
Type: OpenClaw Skill
Name: uplo-hr
Version: 1.0.0
The uplo-hr skill bundle is a legitimate integration for the UPLO HR knowledge management platform. It facilitates searching employee handbooks, policies, and organizational charts using a remote MCP server (@agentdocs1/mcp-server). The instructions in SKILL.md and identity-patch.md are well-aligned with the stated purpose, emphasizing data sensitivity, policy compliance, and proper citation of sources without any evidence of malicious intent, data exfiltration, or harmful prompt injection.
能力评估
Purpose & Capability
The skill is described as an HR knowledge/search integration and the declared capabilities (search_knowledge, search_with_context, directives, etc.) match that purpose. However, registry metadata lists no required env vars while skill.json requires an agentdocs_url and an api_key (MCP token). That mismatch between the top-level Requirements and the packaged config is unexpected and should be resolved.
Instruction Scope
SKILL.md instructs the agent to use UPLO-specific tools (search_knowledge, search_with_context, get_directives, etc.) and to call get_identity_context first for access control. There are no instructions to read local files or unrelated credentials. The guidance emphasizes sensitivity of HR data, which is appropriate. Still, get_identity_context and other calls will surface identity and HR data — ensure that identity context and query results are scoped correctly.
Install Mechanism
The skill package contains no explicit install spec in the registry, but skill.json defines an MCP server launched via: `npx -y @agentdocs1/mcp-server --http`. That implies runtime/npm install activity (fetching and executing code from the npm registry) when the skill is installed or started. Downloading and running remote npm packages is a moderate-to-high risk install mechanism unless the package and publisher are verified; no homepage or published source is provided in the registry metadata to validate provenance.
Credentials
Functionally, requiring an UPLO instance URL and an API key is proportionate for a hosted HR knowledge integration. The concern is the inconsistency: the registry claims no required env vars, but skill.json requires `agentdocs_url` and a secret `api_key`. Requesting an API key is expected, but the absence of those requirements in the top-level metadata is an incoherence that could hide the fact that you'll need to supply a secret to an external endpoint.
Persistence & Privilege
The skill is not flagged with always:true and does not request system-wide configuration changes. Autonomous invocation remains possible (platform default). The main persistence/privilege concern stems from the MCP server model: running an npm-hosted mcp-server process means external code executes on the host and can reach the configured agentdocs_url — review organizational policies on running third-party node packages.
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 uplo-hr - 安装完成后，直接呼叫该 Skill 的名称或使用
/uplo-hr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
uplo-hr 1.0.0 – Initial release
- Launches AI-powered HR knowledge management for centralized, structured search across employee handbooks, policies, benefits, org charts, and onboarding documents.
- Includes tools for knowledge search, context-aware queries, policy directive checks, reporting documentation gaps, and proposing policy updates.
- Provides use-case examples and practical workflow templates for onboarding, leave administration, and compliance requests.
- Emphasizes data sensitivity, access controls, and documentation best practices to support HR accuracy and compliance.
元数据
常见问题
Uplo Hr 是什么？
AI-powered HR knowledge management. Search employee handbooks, org charts, company policies, benefits documentation, and onboarding materials with structured... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 200 次。
如何安装 Uplo Hr？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install uplo-hr」即可一键安装，无需额外配置。
Uplo Hr 是免费的吗？
是的，Uplo Hr 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Uplo Hr 支持哪些平台？
Uplo Hr 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 Uplo Hr？
由 RooJenkins（@roojenkins）开发并维护，当前版本 v1.0.0。
推荐 Skills