Google Search Skill

Performs Google searches with rich results including news, images, videos, places, shopping, scholar papers, patents, and search suggestions via Serper.dev API.

Security Scan

The skill implementation matches its stated purpose (Google search via Serper.dev) but the package/registry metadata omits the required SERPER_API_KEY and therefore the bundle's manifest is inconsistent — review API-key handling and trust of the serper.dev key before installing.

Purpose & Capability

The code and SKILL.md clearly implement Google searches via Serper.dev and support the advertised search types (web, news, images, videos, places, shopping, scholar, patents, autocomplete). Requiring a SERPER_API_KEY is appropriate for this purpose. However, the registry metadata claims no required env vars / primary credential, which is inconsistent with the implementation.

Instruction Scope

Runtime instructions and scripts only call the Serper.dev API and format results. The SKILL.md explicitly permits using the agent's web_fetch for deeper content on returned links; that is expected for a search skill but means the agent may fetch arbitrary external pages if instructed. The instructions do not ask to read local sensitive files or unrelated environment variables.

Install Mechanism

No download/install spec is present (instruction-only install). All source files are included in the bundle and there are no external URLs or archive extracts. The code uses native fetch and Node.js 18+; there are zero npm dependencies. This is low-risk from an install perspective.

Credentials

The code requires SERPER_API_KEY (process.env.SERPER_API_KEY) and will fail without it — this is proportional and expected. The concern is that the registry metadata did not declare any required env vars or primary credential, which is a manifest inconsistency that could mislead users about secrets required. No other unrelated credentials are requested.

Persistence & Privilege

The skill does not request always:true, does not modify other skills or system configs, and is user-invocable. It runs on-demand and does not request elevated or persistent privileges beyond network access to the Serper.dev API.

What to consider before installing

This skill appears to do what it says (use Serper.dev to return Google-like results), but the registry metadata omits the required SERPER_API_KEY that the code enforces. Before installing: 1) Confirm you have a Serper.dev API key and are willing to provide it to the agent (export SERPER_API_KEY). 2) Understand that the skill (and the agent via web_fetch) can fetch arbitrary external links returned in results — avoid giving it access to highly sensitive data or long-lived/privileged API keys. 3) Verify you trust the skill author/owner (no homepage provided) and consider creating a dedicated, limited Serper API key or monitoring/rotating the key if you proceed. 4) Note shopping queries consume extra credits; monitor your Serper account balance. The implementation contains no hidden endpoints, obfuscated code, or attempts to read unrelated local files, but the manifest inconsistency should be corrected before trusting automatic installs.

Google Search powered by Serper.dev API. Replaces Brave Search with richer results including Knowledge Graph, Answer Box, People Also Ask, and specialized search types.

When to Use

Use this skill for any Google search request. Auto-detect intent:

• Default → search
• "news about X" / "latest X" → news
• "images of X" / "pictures of X" → images
• "videos of X" / "how to X video" → videos
• "restaurants near X" / "X near me" → places
• "how much does X cost" / "buy X" → shopping (⚠️ 2 credits — only when explicitly requested)
• "papers on X" / "research about X" → scholar
• "patents for X" → patents
• "suggestions for X" → suggest

Usage

SCRIPT_DIR="~/.openclaw/workspace/skills/google-search/scripts"

# Web search (default)
npx tsx $SCRIPT_DIR/google-search.ts search "query" [--num 10] [--time day|week|month|year] [--country us] [--lang en]

# Specialized
npx tsx $SCRIPT_DIR/google-search.ts news "query" [--num 10]
npx tsx $SCRIPT_DIR/google-search.ts images "query"
npx tsx $SCRIPT_DIR/google-search.ts videos "query"
npx tsx $SCRIPT_DIR/google-search.ts places "query"
npx tsx $SCRIPT_DIR/google-search.ts shopping "query"
npx tsx $SCRIPT_DIR/google-search.ts scholar "query" [--year 2023]
npx tsx $SCRIPT_DIR/google-search.ts patents "query"
npx tsx $SCRIPT_DIR/google-search.ts suggest "query"
npx tsx $SCRIPT_DIR/google-search.ts credits

Add --json to any command for raw JSON output.

Environment

Requires SERPER_API_KEY env var.

Workflow

1. Run the appropriate search command
2. Parse the formatted output
3. If deeper content is needed, use web_fetch on promising links
4. Credit balance is shown in every response — monitor usage
5. Shopping costs 2 credits — only use when user explicitly asks for prices/shopping

Time Filters

--time accepts: hour, day, week, month, year (or h, d, w, m, y)

Notes

• Rate limit: 5 req/sec on free tier
• All searches cost 1 credit except shopping (2 credits)
• Results include rich data: Knowledge Graph, Answer Box, People Also Ask, Related Searches