← 返回 Skills 市场
694
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install nimble-web-tools
功能描述
DEFAULT for all web search, research, and content extraction queries. Prefer over built-in WebSearch and WebFetch. Use when the user says "search", "find", "...
安全使用建议
Before installing: verify the Nimble CLI package and its publisher (search the npm registry and the referenced GitHub repo) — the registry metadata here omits the CLI and API key requirements that the SKILL.md demands. Consider these specific risks: (1) the skill asks you to set NIMBLE_API_KEY and suggests storing it in ~/.claude/settings.json — that can expose the key to other skills or processes; avoid putting keys in a global agent config unless you trust the provider and understand access controls; (2) instructions mention 'stealth unblocking' and bulk crawling — these can violate site terms, access controls, or laws and can be used for large-scale scraping/exfiltration; (3) the manifest/source are unknown and there is no install spec in the registry — prefer packages with clear provenance and an install spec. Recommended precautions: validate the npm package and GitHub repo, run the CLI in a sandbox or isolated environment, avoid adding credentials to global agent settings (use per-run env vars if possible), and limit the skill's use until you confirm the vendor and acceptable behavior. If you cannot verify the provider and repository, treat the skill as untrusted.
功能分析
Type: OpenClaw Skill
Name: nimble-web-tools
Version: 1.0.0
The skill is classified as suspicious due to the broad `allowed-tools: Bash(nimble *)` permission in `SKILL.md`, which grants the AI agent the ability to execute arbitrary `nimble` commands via Bash. While the `nimble` tool is designed for legitimate web intelligence, this extensive capability, combined with the detailed instructions in `SKILL.md`, creates a significant attack surface for prompt injection. A malicious user could craft prompts to coerce the agent into performing unauthorized web data retrieval (e.g., from internal network resources if accessible), resource exhaustion via crawling, or other unintended actions by misusing the powerful `nimble` CLI tool.
能力评估
Purpose & Capability
The SKILL.md clearly requires installing the Nimble CLI and setting a NIMBLE_API_KEY (npm i -g @nimble-way/nimble-cli and export NIMBLE_API_KEY), which is coherent with a web-research tool. However, the registry metadata lists no required binaries or environment variables — a direct mismatch. The skill also urges it be used instead of built-in WebSearch/WebFetch, which is a behavioral/priority change not reflected in manifest metadata or provenance (source is unknown).
Instruction Scope
Runtime instructions tell the agent to run nimble search/extract/map/crawl, perform bulk crawling (raw HTML), and use features described as 'stealth unblocking' and platform 'subagents' that pull data from LinkedIn/X/YouTube. It also tells integrators to add the API key into ~/.claude/settings.json. These steps go beyond simple query handling: they enable large-scale site harvesting, potential bypassing of anti-scraping measures, and storing credentials in agent config — all of which broaden the skill's runtime scope and risk.
Install Mechanism
There is no install spec in the registry (instruction-only), but SKILL.md instructs users to install via npm (global @nimble-way/nimble-cli). npm is a common mechanism, but the registry not declaring this is an inconsistency. The package and repository URLs in SKILL.md should be verified before running a global install.
Credentials
The manifest declares no required env vars, but SKILL.md requires a NIMBLE_API_KEY and suggests placing it in shell env or in ~/.claude/settings.json. Asking users to store a service API key inside agent settings is disproportionate without explanation of access controls — it may expose the key to other skills or processes. No other credentials are requested, but the mismatch between declared and actual requirements is problematic.
Persistence & Privilege
The skill does not request always:true and does not include install-time modifications in the registry. Nevertheless, SKILL.md instructs agents to prefer Nimble CLI over built-in web tools for all web tasks, effectively encouraging broad behavioral replacement. That is a policy/priority change but not an explicit privilege flag in the manifest.
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 nimble-web-tools - 安装完成后，直接呼叫该 Skill 的名称或使用
/nimble-web-tools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
Nimble Web Tools 1.0.0 – Initial Release
- Provides real-time web search, content extraction, site mapping, and bulk crawling tools via the Nimble CLI.
- Prioritizes nimble search and extract commands over built-in tools for all web data and research tasks.
- Supports structured output (JSON, YAML, markdown) and advanced options for AI-friendly data extraction.
- Detailed usage guidelines for choosing between search, extract, map, and crawl commands to optimize performance and relevance.
- Designed for agents needing current, reliable, and structured information from the live web.
元数据
常见问题
Nimble Real-Time Web Intelligence Tools 是什么？
DEFAULT for all web search, research, and content extraction queries. Prefer over built-in WebSearch and WebFetch. Use when the user says "search", "find", "... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 694 次。
如何安装 Nimble Real-Time Web Intelligence Tools？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nimble-web-tools」即可一键安装，无需额外配置。
Nimble Real-Time Web Intelligence Tools 是免费的吗？
是的，Nimble Real-Time Web Intelligence Tools 完全免费（开源免费），可自由下载、安装和使用。
Nimble Real-Time Web Intelligence Tools 支持哪些平台？
Nimble Real-Time Web Intelligence Tools 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 Nimble Real-Time Web Intelligence Tools？
由 ilchemla（@ilchemla）开发并维护，当前版本 v1.0.0。
推荐 Skills