メニュー
Executes Tavily AI search, extraction, crawling, mapping, and deep research via Python CLI. Use when web search, URL content extraction, site crawling, or multi-step research needed.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | tavily-tools |
| user-invocable | false |
| description | Executes Tavily AI search, extraction, crawling, mapping, and deep research via Python CLI. Use when web search, URL content extraction, site crawling, or multi-step research needed. |
Dictum: Command-specific arguments enforce correct invocation.
Execute Tavily AI web operations through unified Python CLI.
[IMPORTANT] search requires --query; extract requires --urls; crawl/map require --url; research requires --query. 1Password injects API key automatically.
| [CMD] | [REQUIRED_ARG] | [PURPOSE] |
|---|---|---|
| search | --query TEXT | AI-powered web search with results |
| extract | --urls URLS | Extract content from one or more URLs |
| crawl | --url URL | Crawl website from base URL |
| map | --url URL | Map website URL structure |
| research | --query TEXT | Multi-step deep research with report |
# Search
uv run .claude/skills/tavily-tools/scripts/tavily.py search --query "Vite 7 new features"
uv run .claude/skills/tavily-tools/scripts/tavily.py search --query "React 19" --topic news
uv run .claude/skills/tavily-tools/scripts/tavily.py search --query "Effect-TS" --search-depth advanced --max-results 20
# Extract content from URLs
uv run .claude/skills/tavily-tools/scripts/tavily.py extract --urls "https://example.com"
uv run .claude/skills/tavily-tools/scripts/tavily.py extract --urls "https://a.com,https://b.com" --format text
# Crawl website
uv run .claude/skills/tavily-tools/scripts/tavily.py crawl --url "https://docs.effect.website"
uv run .claude/skills/tavily-tools/scripts/tavily.py crawl --url "https://nx.dev" --max-depth 3 --max-breadth 50
# Map site structure
uv run .claude/skills/tavily-tools/scripts/tavily.py map --url "https://nx.dev"
uv run .claude/skills/tavily-tools/scripts/tavily.py map --url "https://effect.website" --max-depth 2 --limit 200
# Deep research (multi-step, structured report)
uv run .claude/skills/tavily-tools/scripts/tavily.py research --query "Nx 22 migration strategies"
uv run .claude/skills/tavily-tools/scripts/tavily.py research --query "Effect vs RxJS comparison" --model pro
search: --query TEXT [options]
--query — Search query (required)--topic — Topic: general, news (default: general)--search-depth — Depth: basic, advanced (default: basic)--max-results — Number of results (default: 10)--include-images — Include images in results (flag)--include-raw-content — Include raw HTML (flag)--include-domains — Comma-separated whitelist--exclude-domains — Comma-separated blacklist--time-range — Time filter (e.g., day, week, month, year)--country — Country code for localized resultsextract: --urls URLS [options]
--urls — Comma-separated URLs (required)--extract-depth — Depth: basic, advanced (default: basic)--format — Output: markdown, text (default: markdown)--include-images — Include images (flag)crawl: --url URL [options]
--url — Base URL to crawl (required)--max-depth — Crawl depth (default: 1)--max-breadth — Pages per depth level (default: 20)--limit — Total page limit (default: 50)--allow-external — Follow external links (flag)--select-paths — Comma-separated path filters--instructions — Natural language crawl instructionsmap: --url URL [options]
--url — Base URL to map (required)--max-depth — Map depth (default: 1)--max-breadth — URLs per depth level (default: 20)--limit — Total URL limit (default: 50)--allow-external — Include external URLs (flag)research: --query TEXT [options]
--query — Research question (required)--model — Research agent: mini, pro, auto (default: auto)Commands return: {"status": "success|error", ...}.
| [INDEX] | [CMD] | [RESPONSE] |
|---|---|---|
| [1] | search | {query, results[], images[], answer} |
| [2] | extract | {urls[], results[], failed[]} |
| [3] | crawl | {base_url, results[], urls_crawled} |
| [4] | map | {base_url, urls[], total_mapped} |
| [5] | research | {query, report, sources[]} |
| [VAR] | [REQUIRED] | [DESCRIPTION] |
|---|---|---|
TAVILY_API_KEY | Yes | Tavily API key (1Password injected) |
[ERROR] <status>: <body> and exit 1extract reports per-URL failures in failed[] array (partial success)crawl/map respect --limit to prevent runaway requestsresearch uses extended timeout for multi-step processingUse for Bash 5.2+/5.3 scripts, shell CLIs, entrypoints, CI jobs, cron, text-processing pipelines, ShellCheck remediation, and shell reviews. Enforces strict mode, dispatch-table routing, fork-minimal primitives, atomic I/O, signal-safe cleanup, and pragmatic functional shell patterns.
Enforces C# + LanguageExt functional/ROP style, type discipline, runtime-record effects, Scrutor composition, and module organization standards. Use when writing, editing, reviewing, refactoring, or debugging .cs modules, implementing domain models, sealed DU hierarchies, Eff/Fin/Validation pipelines, or configuring .csproj, analyzers, or DI.
Use for PostgreSQL 18 SQL, migrations, DDL, functions, RLS policies, indexes, query plans, extensions, and SQL embedded in TypeScript via @effect/sql-pg. Enforces set-algebraic queries, schema-level invariants, current PostgreSQL features, extension-first design, tenant security, observability, and migration safety.
Enforces Python + expression style, type discipline, error handling, concurrency, and module organization standards. Use when writing, editing, reviewing, refactoring, or debugging .py/.pyi modules, implementing domain models, ROP pipelines, Protocol-driven services, or configuring pyproject.toml, Ruff, or ty.
Enforces TypeScript + Effect functional/ROP style, type discipline, polymorphic surfaces, and module organization standards. Use when writing, editing, reviewing, refactoring, or debugging .ts/.tsx modules, implementing domain models, Effect services, persistence adapters, or boundary handlers, or configuring TypeScript, Effect, or lint/type-check posture.
Generates and validates project documentation: READMEs, ADRs, changelogs, ARCHITECTURE.md, and code documentation. Use when creating, updating, or reviewing any non-code markdown artifact, README file, architecture decision record, ARCHITECTURE document, CHANGELOG entry, or code documentation (docstrings, XML docs, TSDoc). Activates for: (1) scaffolding README, CHANGELOG, ADR, CONTRIBUTING, ARCHITECTURE, or SECURITY files; (2) writing or reviewing doc comments on exported APIs; (3) auditing documentation coverage or staleness; (4) generating release notes from commit history.