← Back to Skills Marketplace
Tavily Search
by
Evan Rimer
· GitHub ↗
· v7.0.0
· MIT-0
652
Downloads
1
Stars
6
Active Installs
15
Versions
Install in OpenClaw
/install tavily-websearch
Description
Search the web using Tavily's Search API. Returns relevant, accurate results with content snippets, scores, and metadata. Use when the user asks to search th...
Usage Guidance
This skill appears to do exactly what it says: send your search JSON to Tavily's search API using the TAVILY_API_KEY. Before installing, ensure you trust Tavily as a receiver of your queries (they will see your search text and any included options). Keep the API key secret, rotate/revoke it if exposed, and avoid sending sensitive personal data or credentials in the JSON body (the SKILL.md already warns against that). If you need stricter control, disable autonomous invocation for this skill or audit calls that set include_raw_content (which can return full page content). Finally, verify the official Tavily documentation and your organization’s policies on third-party search services.
Capability Analysis
Type: OpenClaw Skill
Name: tavily-websearch
Version: 7.0.0
The skill is a legitimate wrapper for the Tavily Search API. The bash script (scripts/search.sh) safely handles JSON input and communicates only with the official Tavily endpoint (api.tavily.com). The documentation (SKILL.md) provides clear, helpful instructions for the agent without any signs of prompt injection or malicious directives.
Capability Assessment
Purpose & Capability
Name/description, required binary (curl), required env var (TAVILY_API_KEY), and the provided script all align with a web-search API wrapper that posts JSON to https://api.tavily.com/search.
Instruction Scope
SKILL.md and scripts/search.sh only read the provided JSON argument and the TAVILY_API_KEY environment variable and POST them to Tavily's API. There are no instructions to read unrelated files, scan system state, or transmit data to other endpoints.
Install Mechanism
No install spec is present (instruction-only plus a small script). Nothing is downloaded or written by an installer, minimizing disk-write risk.
Credentials
Only a single API credential (TAVILY_API_KEY) is required and declared as the primary credential. That is proportionate for a service that authenticates with a bearer key.
Persistence & Privilege
always:false (default) and no requests to modify agent or system configs are present. Autonomous invocation is allowed by platform default but is not combined with broad privileges here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tavily-websearch - After installation, invoke the skill by name or use
/tavily-websearch - Provide required inputs per the skill's parameter spec and get structured output
Version History
v7.0.0
- Added a new "Data Handling" section with important guidance not to include secrets, credentials, or sensitive personal data in requests.
- No other changes to API, usage, or examples.
v6.0.1
- Added a new "Data Handling" section to the documentation, explaining not to include secrets, credentials, or sensitive data in API requests.
- Clarified to only pass documented fields in the request body; unknown fields are ignored but still sent over the network.
- No functional or behavioral changes to the skill itself.
v6.0.0
- Added explicit homepage and metadata fields in SKILL.md for better discoverability and integration.
- Internal configuration file agents/openai.yaml removed.
- No changes to usage, API, or command structure for users.
v5.0.2
- Removed jq as a required binary; now only curl is required.
- jq is not by default installed, removing friction
- No other changes detected.
v5.0.1
- Added required environment variable (TAVILY_API_KEY) and required binaries (curl, jq) to metadata.
- No changes to functionality or usage instructions.
- Documentation now explicitly lists setup dependencies for easier onboarding.
v5.0.0
No changes detected in this version.
- No updates were made to the skill files or documentation.
- updated the tags
v4.2.0
- Moved the OpenAI agent configuration file from scripts/agents/openai.yaml to agents/openai.yaml.
- No changes to functionality or documentation.
v4.1.0
- Added scripts/agents/openai.yaml to the repository.
- Removed required_env and required_binaries fields from SKILL.md.
v4.0.3
- Fixed typos in example queries (“Lanscape of electric vehichles 2026” → “Landscape of electric vehicles 2026”).
- Added to the metadata required_env:
- TAVILY_API_KEY
required_binaries:
- curl
- jq
v4.0.1
Initial release of the Tavily Search skill.
- Enables web search via Tavily's Search API, returning relevant results with snippets, scores, and metadata
- Supports flexible queries, max results, time ranges, source scoping, and advanced search depth options
- Command-line usage with a `--json` argument mirroring the Tavily Search API request body
- Requires `TAVILY_API_KEY` environment variable and `curl`, `jq` dependencies
- Detailed documentation on all request fields and tips included
v4.0.0
- Added a homepage link to the skill manifest.
- Updated API documentation links to match Tavily's current documentation URLs.
v2.0.0
- Version bump to 2.0.0 to indicate major change
v1.0.2
- Updated requirements to specify only environment variable setup for `TAVILY_API_KEY`
- Added required environment variable and binaries (`curl`, `jq`) to metadata
- Removed dotenv file setup instructions
- Clarified quick start instructions for simplicity
- No functional changes to code detected
v1.0.1
- Simplified authentication: OAuth flow and API key file config removed; now requires TAVILY_API_KEY via env variable or .env file.
- Script usage streamlined: new `--json` flag passes raw Tavily API request body directly.
- Updated documentation: clearer quick start, focused on direct API mapping and typical usage patterns.
- Search depth options reduced: only `basic` and `advanced` modes now documented.
- Description refined to clarify when to use the skill (web search, source lookup, research).
v1.0.0
Major update: Switched from Node.js scripts to a Bash-based search interface with OAuth and detailed documentation.
- Replaced Node.js scripts with a single Bash script (search.sh) for web searches.
- Removed extraction and Node-based search functionality (extract.mjs, search.mjs).
- Introduced OAuth authentication flow; no manual API key setup required on first run.
- Provided comprehensive documentation for API usage, authentication options, and search parameters.
- Added advanced search options, including domain filtering, time ranges, and search depth settings.
Metadata
Frequently Asked Questions
What is Tavily Search?
Search the web using Tavily's Search API. Returns relevant, accurate results with content snippets, scores, and metadata. Use when the user asks to search th... It is an AI Agent Skill for Claude Code / OpenClaw, with 652 downloads so far.
How do I install Tavily Search?
Run "/install tavily-websearch" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Tavily Search free?
Yes, Tavily Search is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Tavily Search support?
Tavily Search is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Tavily Search?
It is built and maintained by Evan Rimer (@evanydl); the current version is v7.0.0.
More Skills