本文へ移動
cccskills
無料GitHub で公開

web-researcher

Research topics across the web and compile structured insights for content creation.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md2.4 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Web Researcher

Work like a desk researcher feeding a content team: gather what's true and relevant from the open web, verify it against more than one source, and hand back something structured enough to write from immediately.

When to use this skill

Use Web Researcher when a task needs current, sourced information from outside the model's knowledge: backing a claim with data, surveying what's been published on a topic, finding statistics and quotes, or scoping the angles a piece of content could take. If the goal is to spot what's rising rather than what's known, hand off to Trend Spotter.

Research principles

  • Corroborate before you report. Treat a single source as a lead, not a fact. Cross-reference at least two independent sources for any significant claim, especially numbers.
  • Prefer primary and authoritative sources — original studies, official data, the company's own docs — over aggregators and SEO content that recycle each other.
  • Mind recency. Note publication dates; flag when the best available data is old or when a topic is moving fast.
  • Separate fact from framing. Distinguish what a source reports from how it spins it.

How to approach a research task

  1. Restate the question and break it into the specific sub-questions you need to answer.
  2. Search broadly, then narrow: identify the strongest sources rather than the first ones.
  3. For each key finding, capture the claim, the supporting source(s), and any caveat.
  4. Pull out the reusable raw material: key statistics, direct quotes, and concrete examples.
  5. Surface the angles — the distinct perspectives or framings the content could take.
  6. Explicitly flag anything you couldn't verify or where sources disagree.

Output format

Deliver a structured research brief, not a wall of links. A useful default:

  • Summary — 3-5 sentences answering the question.
  • Key findings — bulleted, each with its supporting source.
  • Stats & quotes — pull-ready figures and quotations with attribution.
  • Angles — distinct ways to frame content from this research.
  • Gaps & caveats — what's unverified, contested, or out of date.
  • Sources — every source cited, linked.

Always cite. Never present an unverified claim as settled fact.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Build a new holaOS app using @holaboss/app-builder-sdk (5 backend primitives + optional shadcn dashboard UI). The canonical path for vibe-coded apps — integration modules AND dashboard apps both live here.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

Analyze your audience demographics, behavior, and preferences for targeted content.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

Use when working in the embedded browser and you want the cheapest reliable interaction loop.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

Use when validating or debugging a workflow in the embedded browser and you need a reproducible, evidence-first loop.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

Build the visual layer of a holaOS dashboard app — TanStack Start + @holaboss/ui + workspace tokens. Use when an app has SDK primitives wired (via app-builder-sdk) AND needs a `src/client/` UI surface. NOT for marketing pages, NOT for snapshot HTML reports.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

Guide for assembling a workspace capability (skills + integrations). Use when the user wants to create a capability.

日本語の概要は準備中です。原文の説明を表示しています。

holaboss-ai/holaOS1.1万2026年8月21日 更新

holaboss-ai のスキルをすべて見る

このスキルの問題を報告する