AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Modsearch

skill-liustack-modsearch-modsearch · by liustack

Bridge web search capability for LLM workflows. Use when user asks for latest info, external facts, or source links and the active model/toolchain lacks direct search ability.

— No reviews yet
0 installs
22 views
0.0% view→install

Install

$ agentstack add skill-liustack-modsearch-modsearch

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-liustack-modsearch-modsearch)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
○ 6mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Modsearch? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

ModSearch - Search Bridge Skill

Use this skill when:

  • User asks for web search, latest updates, or source-backed answers
  • Current model/toolchain cannot perform direct search reliably
  • You need structured search results before downstream summarization/fetch

Prerequisites

modsearch --version

If using the default provider (gemini-cli), ensure Gemini CLI is installed and authenticated:

gemini --version

If missing:

npm install -g @google/gemini-cli
gemini

Or run directly with npx (no global install required):

npx @google/gemini-cli          # launch interactive auth

Command

modsearch -q ""
# or via npx
npx @liustack/modsearch -q ""

Optional:

modsearch -q "" -o  -p  -m  --max-results  --prompt ""

Workflow

  1. Run modsearch with the user query.
  2. Parse JSON output.
  3. Use summary and items as evidence for reasoning or for selecting URLs to fetch.
  4. If uncertainty is non-empty, explicitly communicate ambiguity to the user.

Output Contract

  • summary: high-level synthesis of search findings
  • items: normalized list (title, url, snippet, source, published_at, relevance)
  • uncertainty: caveats, missing data, or confidence risks

Detailed schema: references/output-schema.md

Failure Handling

  • If provider command fails (missing auth, quota, network, binary not found), report exact error and suggest provider setup checks.
  • If JSON is partially malformed, keep rawText and continue with best-effort extraction.

Tavily Provider

To use the Tavily search provider, set the TAVILY_API_KEY environment variable:

export TAVILY_API_KEY="tvly-YOUR_API_KEY"

Then run with --provider tavily (or -p tavily):

modsearch -q "" -p tavily
# or via npx
npx @liustack/modsearch -q "" -p tavily

Get your API key at https://app.tavily.com (1,000 free credits/month).

Implementation Note

v1 uses Gemini CLI as the default provider (gemini -p with JSON output mode). The architecture is provider-extensible and can support any search-capable model or service provider in future versions.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.