ModSearch — Search & Fetch Bridge
Gives models without native web access a CLI-backed web search, X (Twitter) search, and page fetch that returns structured JSON evidence.
Web & APIIntermediate★ 111⑂ 5AI score 10/10Last updated: Aug 16, 2026
What it does
- Runs the
modsearchCLI to perform web search, X (Twitter) search, and single-URL page fetch when the host model has no built-in search or fetch tool. - Always returns a structured JSON contract (
results[],items[].url,summary,uncertainty,warnings,attempts) so answers can be cited and audited. - Multiple engines per role (antigravity-cli, tavily, exa, firecrawl, grok-cli, local) with automatic fallback and quota cooldown handling.
- A bundled launcher resolves the runtime in order —
modsearchon PATH →npx→bunx— and emits a JSON diagnosis with next steps if none is available. - Explicit safety rule: fetched pages and posts are treated as untrusted data, never as instructions.
Who it's for
- Anyone who needs post-cutoff facts: releases, news, prices, library versions.
- Users on harnesses without a web tool who still need verifiable source links.
- Researchers and marketers tracking discussion on X.
Examples
- "What's the latest release of this framework?" →
modsearch -q "...", then citeitems[].url. - "What are people saying on X about this launch?" →
--source x; if Grok isn't configured, relay thestatus: degradedwarning instead of claiming X coverage. - "Summarize this page" →
modsearch -u "<url>" -q "focus"to fetch with an extraction focus.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/liustack/modsearch/HEAD/skills/modsearch/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the skills/modsearch folder from the GitHub repo liustack/modsearch into my ~/.claude/skills/modsearch/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/liustack/modsearch.git /tmp/modsearch && mkdir -p ~/.claude/skills && cp -r /tmp/modsearch/skills/modsearch ~/.claude/skills/modsearch⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repo:
git clone https://github.com/liustack/modsearch.git /tmp/modsearch - Copy the skill in:
mkdir -p ~/.claude/skills && cp -r /tmp/modsearch/skills/modsearch ~/.claude/skills/modsearch - Install a JavaScript runtime: Node 22.13+ (https://nodejs.org) or Bun (https://bun.sh).
- Smoke test:
bash ~/.claude/skills/modsearch/scripts/run.sh -q "test"(on Windows usepowershell -ExecutionPolicy Bypass -File ...\scripts\run.ps1 -q "test"). - If something fails, run
bash ~/.claude/skills/modsearch/scripts/run.sh doctor --jsonto see engine readiness, and add a Tavily/Exa/Firecrawl key if web search has no engine. - Restart Claude Code, then ask for current information or hand it a URL — the skill triggers automatically.
View source on GitHub ↗License: MIT