← 返回 Skills 市场
Uplo Consulting
作者
RooJenkins
· GitHub ↗
· v1.0.0
· MIT-0
234
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install uplo-consulting
功能描述
AI-powered consulting knowledge management. Search engagement records, methodology frameworks, deliverable templates, and best practices with structured extr...
安全使用建议
This skill appears to be a legitimate connector to an UPLO consulting knowledge service, but there are packaging inconsistencies and a moderate installation risk to consider. Before installing: 1) Verify the publisher and contents of the npm package @agentdocs1/mcp-server (review its code or use a vetted internal mirror). 2) Confirm the correct UPLO instance URL and create an API key scoped to the minimum required permissions (read-only search, and separate keys for logging/writes if needed). 3) Treat the API key as a secret and do not reuse highly privileged cloud/service keys. 4) Because the registry listing omitted the required credentials while skill.json requires them, expect the agent to request or need an API key — do not supply credentials until you confirm the endpoint and package. 5) If possible, run the MCP server in an isolated environment or behind network controls and monitor traffic to ensure data remains within your org. If you cannot verify the npm package or do not want to provide an external API key, do not install.
功能分析
Type: OpenClaw Skill
Name: uplo-consulting
Version: 1.0.0
The skill bundle provides a legitimate interface for a consulting knowledge management platform (UPLO). The instructions in SKILL.md and identity-patch.md are well-aligned with the stated purpose of retrieving engagement records and methodology frameworks, and the MCP configuration in skill.json uses standard patterns for connecting to an external API via the @agentdocs1/mcp-server package.
能力评估
Purpose & Capability
The skill's name/description and SKILL.md consistently describe a consulting knowledge retrieval tool (search_knowledge, search_with_context, get_directives, export_org_context, etc.) and the declared capabilities align with that purpose. However, the registry metadata provided earlier lists no required credentials while the included skill.json clearly requires an UPLO instance URL and an API key — a packaging/manifest inconsistency that could cause the agent to request secrets unexpectedly.
Instruction Scope
SKILL.md instructs only on querying the UPLO knowledge service and logging conversation metadata (e.g., get_identity_context, get_directives, search_knowledge, log_conversation). It does not ask the agent to read unrelated local files, environment variables, or exfiltrate data to unexpected endpoints. The identity-patch biases the agent to consult UPLO first, which is functionally expected for this skill but should be noted as behavioral biasing.
Install Mechanism
There is no explicit 'install spec' in the registry, but skill.json and README show an MCP runtime that runs via 'npx @agentdocs1/mcp-server --http' (i.e., fetching and executing an npm package at runtime). Using npx to fetch/execute a remote npm package is a moderate risk (remote code execution at install/run time). The package origin (@agentdocs1) is not a widely known system package in this metadata — verify the npm package, its publisher, and contents before running.
Credentials
The skill requires an external service URL and an API key (agentdocs_url / api_key) to function. Those are reasonable for a knowledge-base integration, but the earlier registry summary claiming 'no required env vars / primary credential: none' contradicts the included skill.json and README. That mismatch is concerning because it means the skill may prompt for or expect sensitive credentials even though the registry listing did not declare them. Ensure the API key is least-privilege and that you understand what scopes the UPLO MCP token grants (search, read/export, write/logging).
Persistence & Privilege
The skill is not set to always:true and does not request elevated platform privileges or modify other skills. Autonomous invocation is allowed (default) but not by itself a red flag here. The skill does include an identity patch that biases use toward UPLO data, which is expected for a knowledge connector.
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-consulting - 安装完成后，直接呼叫该 Skill 的名称或使用
/uplo-consulting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 uplo-consulting.
- Enables AI-powered consulting knowledge management with structured retrieval.
- Search and extract engagement records, frameworks, templates, and best practices.
- Provides tooling for proposal development, staffing, onboarding, and lessons learned.
- Includes directives and reporting functions to support firm operations and compliance.
元数据
常见问题
Uplo Consulting 是什么？
AI-powered consulting knowledge management. Search engagement records, methodology frameworks, deliverable templates, and best practices with structured extr... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 234 次。
如何安装 Uplo Consulting？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install uplo-consulting」即可一键安装，无需额外配置。
Uplo Consulting 是免费的吗？
是的，Uplo Consulting 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Uplo Consulting 支持哪些平台？
Uplo Consulting 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 Consulting？
由 RooJenkins（@roojenkins）开发并维护，当前版本 v1.0.0。
推荐 Skills