Tavily Web Search

Search the web via the Turing Tavily proxy API.

Usage

python3 ~/.openclaw/skills/turing-tavily-web-search/scripts.py '<JSON>'

Request Parameters

| Param | Type | Required | Default | Description | |---|---|---|---|---| | query | str or list[str] | yes | — | Search query, or list of queries for batch search | | max_results | int | no | 10 | Max results per query (1–20) | | max_tokens_per_page | int | no | 1024 | Max tokens extracted per result page | | search_domain_filter | list[str] | no | — | Restrict results to specific domains (max 20) |

Response Fields

| Field | Type | Description | |---|---|---| | answer | str | AI-generated summary (if available) | | results[].title | str | Page title | | results[].url | str | Page URL | | results[].content | str | Content snippet | | results[].publishedDate | str | Publication date (if available) |

Examples

# Basic search
python3 ~/.openclaw/skills/turing-tavily-web-search/scripts.py '{"query": "latest AI news"}'

# Limit results
python3 ~/.openclaw/skills/turing-tavily-web-search/scripts.py '{"query": "latest AI news", "max_results": 5}'

# Domain filter
python3 ~/.openclaw/skills/turing-tavily-web-search/scripts.py '{"query": "transformer architecture", "search_domain_filter": ["arxiv.org", "github.com"]}'

Configuration

Set in openclaw.json under skills.entries.turing-skills.env:

| Variable | Required | Description | |---|---|---| | TURING_API_KEY | yes | Bearer token (sk-...) | | TURING_CLIENT | yes | Client identifier | | TURING_ENVIRONMENT | yes | Environment name | | TURING_API_BASE | no | API base URL (default: https://live-turing.cn.llm.tcljd.com) |