Does this skill require external API keys?

Yes, to use the web search and code search features, you will need API keys for Tavily and Exa respectively. Standard URL scraping may also require specific tool setups like the GitHub CLI.

Where are the gathered artifacts stored?

Artifacts are saved to a designated OUTPUT_DIR, which defaults to .cwf/projects or a workspace-local fallback directory to ensure they can be referenced by downstream tasks.

What platforms can the Gather skill extract data from?

The skill supports automated extraction and markdown conversion from Google Docs, Slack archives, Notion sites, GitHub repositories, and general web URLs.

Is the web scraping secure?

The skill includes a mandatory URL safety pre-check that blocks access to localhost, loopback, and private IP ranges unless explicitly overridden by the user.

How does the local codebase exploration work?

In --local mode, the skill uses tools like Glob and Grep to scan your repository for specific patterns and generates a structured markdown summary including key files and architecture.

Context Gatherer

Name: Context Gatherer
Author: corca-ai

bycorca-ai

•

Web Scraping & Data Collection

Acquires and stabilizes information from URLs, web searches, and local codebases into reusable markdown artifacts for AI reasoning.

The Gather skill is a powerful utility designed to solve context fragmentation by converting scattered information sources into stable, local markdown files. It features intelligent auto-detection for diverse platforms including Google Docs, Slack threads, Notion pages, and GitHub repositories, alongside advanced web search capabilities via Tavily and Exa. By normalizing heterogeneous data into local artifacts, it ensures that subsequent AI reasoning phases operate on a consistent, versionable, and reliable context rather than transient links or volatile web content.

Key Features

01Standardized metadata capture including source URLs and safety pre-checks

02Automated local codebase exploration with structured markdown summaries

03Advanced web, news, and deep research searching powered by Tavily

04Technical and code-specific search integration via Exa API

0573 GitHub stars

06Multi-source URL auto-detection for Google Docs, Slack, Notion, and GitHub

Use Cases

01Consolidating project requirements from Slack threads and Google Docs into a local development environment

02Generating an architectural overview of a local codebase to provide better context for refactoring tasks

03Performing deep technical research for new library implementations using web and code searches

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add corca-ai/claude-plugins gather

For use in Claude.ai and ChatGPT

Download Skill