skills.rest
TeXmeijin
TeXmeijinCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ggg

Searches the web with current-year queries and cites official sources for verified answers.

Updated May 20, 2026
Search Tags:#research#web search#google search#fact checking#source citation#freshness verification
One-click install
npx skills add https://github.com/TeXmeijin/agent-skills --skill ggg-texmeijin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ggg
Source: https://github.com/TeXmeijin/agent-skills/tree/main/.apm/skills/ggg
Command: npx skills add https://github.com/TeXmeijin/agent-skills --skill ggg-texmeijin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? LLM training data goes stale, so answers about current versions, recent releases, or fast-moving topics can be outdated or wrong. This Skill forces every answer to be grounded in live web search results with cited sources instead of relying on memorized knowledge. ## Core Features & Use Cases - Mandatory Web Search: Always uses the WebSearch tool rather than answering from model knowledge, even when the answer seems known. - Freshness-First Queries: Adds the current year to queries and prioritizes official documentation, release notes, and official blogs before falling back to Stack Overflow, Zenn, Reddit, or GitHub. - Doubt Verification: Triggers on challenges like "本当に？" or "それ最新？" to fact-check a previous answer and pinpoint exactly what was inaccurate or outdated. - Use Case: A user asks whether a library API changed in the latest release. The Skill searches official release notes and docs in both English and Japanese, then answers with the conclusion first and source URLs listed at the end. ## Quick Start Ask the agent to search the web for the latest information on a topic, for example by saying "ググって" or asking it to verify whether a previous answer is still current.

Frequently Asked Questions about ggg

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make an AI agent always search the web before answering?▼

Use a skill that mandates web search for every query rather than allowing answers from model knowledge. This Skill triggers on explicit requests like "search for this" or doubt phrases like "is that current?" and always returns cited sources.

How to verify if an AI answer is outdated or wrong?▼

Challenge the answer with phrases like "本当に？" or "それ最新？" to trigger a verification search. The Skill re-searches official sources and identifies exactly which part of the previous answer was inaccurate or stale.

What sources does web search prioritize for technical questions?▼

Official documentation, official blogs, and release notes are checked first. If those are insufficient, the search expands to Stack Overflow, Zenn, personal blogs, Reddit, and GitHub Issues or Discussions.

Does web search work for both English and Japanese queries?▼

Yes, the Skill searches in both English and Japanese, prioritizing English for technical topics. Queries include specific error messages, version numbers, and the current year to filter out stale results.

When should I not rely on web search for answers?▼

Web search adds latency and is unnecessary for stable, well-established knowledge like fundamental algorithms or math. It is most valuable for fast-moving topics such as library versions, recent releases, and breaking changes.

Related Skills

Explore More Education & Research Skills→
firecrawlfirecrawl
72

firecrawl-research-index

Retrieve research papers via Firecrawl semantic search with in-body verification.

Official
Advanced
nebinfranebinfra

neb-research

Synthesize domain knowledge into a cited reference document with key findings and gaps.

Official
Intermediate
jbrowning24jbrowning24

peer-review

Verify external code-review findings against the actual codebase.

Community
Advanced
javiruujaviruu

viru-air-context

Rebuilds working context for the Viru Air repo using canonical documentation references.

Community
Basic