← Back to Skills Marketplace
Tavily Best Practices
by
Evan Rimer
· GitHub ↗
· v1.0.0
· MIT-0
347
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install tavily-bestpractices
Description
Build production-ready Tavily integrations with best practices baked in. Reference documentation for developers using coding assistants (Claude Code, Cursor,...
Usage Guidance
This skill is documentation for the Tavily SDK and integrations (search, extract, crawl, research). It is instruction-only and does not itself request or store secrets, but example code shows using TAVILY_API_KEY and other service keys (OpenAI, Cohere, MongoDB URIs). Before copying examples into real projects: only use real API keys with the official Tavily SDK, avoid pasting secrets into public places, and review any integration that stores web results in databases (e.g., save_foreign=True) to ensure you control where data and credentials are sent. If you want the agent to make live Tavily calls, plan to provide a Tavily API key deliberately and only to trusted code paths.
Capability Analysis
Type: OpenClaw Skill
Name: tavily-bestpractices
Version: 1.0.0
This skill bundle is a comprehensive documentation and reference guide for the Tavily Search API, designed to help developers integrate web search and research capabilities into AI agents. It contains legitimate code examples for Python and JavaScript SDKs, as well as detailed integration guides for popular frameworks like LangChain, Pydantic AI, and CrewAI. No evidence of malicious logic, data exfiltration, unauthorized execution, or prompt injection was found across the provided files (SKILL.md and the references directory).
Capability Assessment
Purpose & Capability
The name and description promise Tavily integration best practices and the files are a detailed SDK and integration guide for search, extract, crawl, map, and research — which matches the stated purpose. There are no unexpected binaries or privileged hooks requested by the skill itself.
Instruction Scope
The SKILL.md and reference files are instructional samples and code snippets (pip/npm install, client usage, async patterns, crawling and research flows). They do not instruct the agent to read arbitrary local files or exfiltrate data, but many examples reference environment variables, external services, and connecting to external resources (e.g., MongoDB URIs, OpenAI/Cohere keys, proxies). Those are example usage patterns rather than instructions the skill will autonomously execute, but users should be aware code snippets may prompt developers to provide credentials or connection strings when they copy them into real applications.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to execute. Nothing is written to disk by the skill itself — lowest install risk.
Credentials
The skill declares no required environment variables, which is consistent for a docs-only skill. However the documentation repeatedly shows TAVILY_API_KEY (and in integration examples references OPENAI_API_KEY, CO_API_KEY, MongoDB connection URIs, etc.). These example credentials are expected for the demonstrated integrations but are not required by the skill itself — users should not supply keys to this skill unless they intend to run the corresponding client code.
Persistence & Privilege
The skill does not request persistent presence (always:false) and does not attempt to modify other skills or system-wide settings. Autonomous invocation is enabled by default but is not coupled with any additional privileges or credential requests here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tavily-bestpractices - After installation, invoke the skill by name or use
/tavily-bestpractices - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
tavily-best-practices 1.0.0
- Initial release with best-practice reference documentation for Tavily’s Python and JavaScript SDKs.
- Covers web search, content extraction, site crawling, URL discovery, and end-to-end AI research integration.
- Includes quick-start code examples and guides to key parameters for each workflow.
- Provides links to detailed references for SDK usage, query optimization, async patterns, and integrations with popular frameworks.
Metadata
Frequently Asked Questions
What is Tavily Best Practices?
Build production-ready Tavily integrations with best practices baked in. Reference documentation for developers using coding assistants (Claude Code, Cursor,... It is an AI Agent Skill for Claude Code / OpenClaw, with 347 downloads so far.
How do I install Tavily Best Practices?
Run "/install tavily-bestpractices" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Tavily Best Practices free?
Yes, Tavily Best Practices is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Tavily Best Practices support?
Tavily Best Practices is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Tavily Best Practices?
It is built and maintained by Evan Rimer (@evanydl); the current version is v1.0.0.
More Skills