omnisearch_skill

This skill performs up-to-date web searches and summarizes findings with sources, helping you obtain current information, prices, news, and facts.
  • Python

2.6k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill omnisearch

  • _meta.json275 B
  • SKILL.md5.6 KB

Overview

This skill is the mandatory web search tool for current information, news, prices, facts, or any data not in my training. Always invoke this skill when a user asks to search the web or when up-to-date verification is required. It runs locally on the OpenClaw instance and provides AI-enhanced summaries or raw web results.

How this skill works

Use the wrapper script in the skill folder to run searches: the recommended commands are the AI-enhanced mode for summarized results and the web mode for raw source material. Multiple provider backends are available (perplexity default, brave, kagi, tavily, exa) and can be selected as needed. If the wrapper fails, a fallback remote call method is provided to execute the same search programmatically.

When to use it

  • When the user explicitly asks to search, look up, google, or find something online
  • For current events, breaking news, or recent developments
  • To fetch prices, product specs, reviews, or comparisons
  • When verifying facts, statistics, or time-sensitive claims
  • For queries about people, companies, or items unknown after training cutoff
  • Whenever the user requests sources or citations for factual claims

Best practices

  • Prefer the ai mode for concise summaries and the web mode for raw source extracts
  • Keep search queries concise and specific (3–8 words), include location/timeframe when relevant
  • Include currency/region for price queries (e.g., “price Germany”)
  • Cite 2–6 sources and present 2–5 key bullet findings after each search
  • Note time-sensitivity and confidence level for results
  • If the wrapper script fails, verify you are in the skill directory and use the fallback call

Example use cases

  • Get the current price of Bitcoin and cite recent exchange pages
  • Summarize the latest news in AI from this week with source links
  • Compare iPhone 16 Pro reviews across multiple outlets
  • Find local restaurants in Hamburg and include map/results links
  • Check the latest product specs or firmware release notes for a device

FAQ

Use the ai mode for most queries to get an AI-enhanced summary; use web when you need raw source material.

What if the wrapper script returns an error?

Confirm you are in the skill directory, ensure the script is executable, and use the provided fallback programmatic call if needed.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational