everysearch — Claude Code skill: search and setup
everysearch is a Claude Code skill for finding current sources and checking facts through different retrieval methods. It activates for requests that need recent information or material from a specific public page.
Apache-2.0 ★ 681 updated 2026-10-24
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a claude-codeWhat everysearch does
everysearch picks a retrieval method to match the question. It can search the web, semantic results, academic material and forums, as well as read pages and crawl sites. For a concrete fact such as a current price or version, it uses crosscheck to compare distinct retrieval paths.
For a known URL, the skill uses page-reading or crawl tools instead of searching for that same page again. Factual conclusions should cite the original sources returned by search. The MCP tool everyinfra_search is required; its live schema determines which modes and parameters are available. An empty result is not evidence that the information does not exist.
What it helps with
- find recent information across the web
- compare sources to verify a price or version
- retrieve content from a public URL
- locate academic material or forum discussions
Example prompts for the agent
- Find recent sources on this library's current version and include links.
- Check this service's price using distinct search methods, then compare the sources.
- Read this public page and summarize its factual claims with citations.
How to install everysearch
1Skills CLI (Vercel)
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a claude-code2Manually: put the files in place
~/.claude/skills/everysearch/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/everysearch/
1Skills CLI (Vercel)
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a codex2Manually: put the files in place
.agents/skills/everysearch/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a cursor2Manually: put the files in place
.cursor/skills/everysearch/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install everyinfra/everyinfra-agent-plugins plugins/everyinfra/skills/everysearch2Manually: put the files in place
.github/skills/everysearch/SKILL.md1Skills CLI (Vercel)
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a gemini-cli2Platform CLI
gemini skills install https://github.com/everyinfra/everyinfra-agent-plugins --path plugins/everyinfra/skills/everysearchManually: put the files in place
.devin/skills/everysearch/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about everysearch
How do I install everysearch in Claude Code?
Run npx skills add everyinfra/everyinfra-agent-plugins --skill everysearch -a claude-code. The skill file is at ~/.claude/skills/everysearch/SKILL.md.
What does everysearch do?
It retrieves current information through several methods, reads pages from URLs, and compares sources to verify specific claims. Factual conclusions use citations to original sources.
What does everysearch need to work?
The MCP tool everyinfra_search must be available. Its current schema defines the supported retrieval modes and parameters.
Source text
SKILL.md · Apache-2.0
--- name: everysearch description: Use EverySearch for current web, semantic, academic, forum, page-reading, crawling, and cross-check retrieval. Use when a task needs up-to-date sources, multiple retrieval mechanisms, cited factual verification, or content from a specific public URL or site. --- # EverySearch Use the MCP tool `everyinfra_search`. Its live input schema is the source of truth for available `tool` values and parameters. - Use the narrowest retrieval mechanism that matches the user's question. - Use `crosscheck` for a concrete claim such as a current version, price, policy, availability, or API behavior. It is designed to compare mechanism-distinct retrieval paths. - Use page-reading or crawl tools for known URLs instead of searching for the same page again. - Keep query and result limits proportional to the task. Empty or failed results must not be presented as proof that nothing exists. - Cite the original sources returned by the search result when producing factual conclusions. - Read the returned `billing` state. Failed or empty calls may be uncharged, but do not infer that without the response. Do not silently replace semantic retrieval with keyword retrieval when result meaning matters. If the requested tool is not in the current MCP schema, state that it is unavailable.
© everyinfra · Apache-2.0 · everyinfra/everyinfra-agent-plugins