Skip to content

sandbaseai/sandbase-skills / research/tavily-deep-research

tavily-deep-research

Verified

Advanced web search, content extraction, and site mapping through Tavily via SandBase. Use when asked for web research, URL content extraction, full article reading, news search, or site structure discovery.

FileKindSize
SKILL.mdskill md2.7 KB
references/sandbase-api-map.mdmarkdown985 B
---
name: tavily-deep-research
description: Advanced web search, content extraction, and site mapping through Tavily via SandBase. Use when asked for web research, URL content extraction, full article reading, news search, or site structure discovery.
---

# Tavily Deep Research

Advanced web search and content extraction through SandBase's Tavily integration. Search the web with depth control, extract clean content from URLs, and discover site structures. Read [the API map](references/sandbase-api-map.md) before selecting a capability.

## Call SandBase capabilities

Use the capability identifiers below as discovery hints, not MCP tool names. Find the matching endpoint with `sandbase_discover(q: "<provider and capability>")`; use its returned `name` in `sandbase_inspect(name: "<returned name>")`. Read `inputSchema`, pricing, and `execute_as`, then call `sandbase_run` using `execute_as.arguments.name` and schema-defined `arguments`. If a `run_id` is returned, poll `sandbase_run_get(run_id: "<returned run_id>")` within the task budget until `completed` or `failed`; report pending or failed runs without resubmitting them automatically.

## Operating principles

- Use Tavily for current, real-time web information that the model's training data may not cover.
- Cite sources with URLs for every factual claim.
- Use appropriate search depth: basic for quick checks, advanced for thorough research.
- Extract full content only when necessary — search snippets are often sufficient.

## Workflow

### 1. Search the web

Use `tavily_search` with parameters:
- `search_depth`: "basic" for quick validation, "advanced" for thorough research
- `topic`: "general", "news", or "finance" to focus results
- `days`: limit to recent results (e.g., 7 for past week)
- `max_results`: 5-20 depending on coverage needs
- `include_domains` / `exclude_domains`: filter by source

### 2. Extract content

Use `tavily_extract` to get clean, readable content from specific URLs.
- Use when search snippets aren't enough and full article text is needed.
- Works on most public web pages — not paywalled content.

### 3. Map site structure

Use `tavily_map` to discover all pages on a website.
- Use for competitor site analysis or content auditing.
- Returns site structure and page list.

## Output

Return: search results with source URLs, extracted content summaries, site maps when relevant, and confidence indicators.

## Example tasks

- "Search the web for the latest news about [topic] in the last 7 days."
- "Extract the full content from this article: [URL]."
- "Research [topic] — find 10 authoritative sources and summarize key findings."
- "Map all pages on [website URL] to understand their content structure."
- "Find recent news about [company] and summarize the key developments."
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall