# eze-is/web-access

给 Claude Code 装上完整联网能力的 skill：三层通道调度 + 浏览器 CDP + 并行分治

Repository: https://github.com/eze-is/web-access
Canonical: https://ross.abutalabs.com/products/web-access
Language: JavaScript
License Family: other
Last push: 2026-08-19T01:36:35+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 52, longevity 12
- inputs: {"age_days": 168, "days_push": 15, "days_rel": 110, "gap_med": null, "n_releases_24m": 1}
- flags: young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8745, forks 628 (observed 2026-08-28T04:10:25.485817+00:00)

## What it is
An Agent Skill that gives AI coding agents (Claude Code, Cursor, Gemini CLI, etc.) full web access capabilities: three-tier channel dispatch across WebSearch/WebFetch/curl/Jina/CDP, plus browser automation via Chrome DevTools Protocol. It supports parallel sub-agent execution, per-domain site experience accumulation, and media extraction from live pages.

## Use cases
- let my AI agent browse and interact with dynamic web pages
- automate browser actions with the user's logged-in Chrome session
- scrape content from pages that block plain HTTP fetches
- run parallel web research tasks across multiple tabs
- extract images or video frames from a webpage for analysis
- search my local browser bookmarks and history for internal sites
- give Claude Code web fetching and browsing capabilities

## When to choose
- you use an agent that supports SKILL.md and need robust web access beyond built-in WebSearch/WebFetch
- you need CDP-driven browser automation that reuses existing login sessions
- your workflows involve dynamic pages, file uploads, or video frame capture
- you want multi-target web tasks parallelized with tab-level isolation

## When to avoid
- you need a standalone scraping library for a non-agent pipeline
- you require a formally licensed product (no license is specified)
- your environment cannot run a local browser or CDP proxy
- you only need simple static HTTP fetching already covered by built-in tools

## Facets
- artifact type: plugin
- maturity: active
- function: web-scraping, http-client, agent-framework, browser-extension, search-engine
- domain: web-development, developer-tools, crawlers
- platform: cross-platform, cli
- tags: agent-skill, claude-code, cdp, browser-automation, web-access, skill-md, chrome-devtools-protocol, ai-agents, automation, nodejs

## Member repositories
- eze-is/web-access (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:25.485817+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:25:17.162136+00:00, confidence not recorded.
  - readme: https://github.com/eze-is/web-access (fetched 2026-08-28T04:10:25.485817+00:00, sha 0f4e81fcf0d0)
- Data as of 2026-08-30T08:39:29.467469+00:00.