code-yeongyu

Web Fetch — AI skill for Claude Code

AI community

LLM-neutral skill that fetches a URL to a temp file so agents can pipe the path through rg / jq / awk instead of dumping the whole page into context.

How to install Web Fetch

This entry records only its repository, not the path inside it, so there is no exact command to give. Open code-yeongyu/web-fetch and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Web Fetch does

LLM-neutral skill that fetches a URL to a temp file so agents can pipe the path through rg / jq / awk instead of dumping the whole page into context. Same shape as pi-webfetch / opencode webfetch. Single-file Python 3 stdlib. Works on macOS, Linux, Windows.

Alternatives in AI

Qiaomu Markdown Proxy — Fetch any URL as clean Markdown via proxy services (r.jina.ai / defuddle.md) or built-in scripts 426 ★
Devo — Model-neutral agent desktop/runtime for private, enterprise, and OpenAI-compatible / Anthropic-compatible mode 316 ★
Temp Doc — Manage temporary documentation created during task execution - prevent git pollution from AI-generated checkli 64 ★

README

web-fetch

LLM-neutral skill for fetching a URL and writing the body to a temp file so an agent can pipe the path through `rg` / `jq` / `awk` instead of dumping the whole page into the conversation.

Same shape as [`pi-webfetch`](https://github.com/code-yeongyu/pi-webfetch) and [`opencode/webfetch`](https://github.com/sst/opencode), packaged as a standalone skill that any Bash-capable agent (Claude Code, OpenCode, pi, hermes, openclaw) can load.

Install

git clone https://github.com/code-yeongyu/web-fetch ~/.agents/skills/web-fetch

That is it. The script is single-file Python 3 stdlib; no `pip install` needed. `curl` is preferred but optional (urllib fallback).

Symlink for active development

ln -s /path/to/your/clone ~/.agents/skills/web-fetch

Other agents

Claude Code / OpenCode: drop the directory under ~/.agents/skills/ and the skill auto-registers via the name + description in the frontmatter.
pi (~/.senpi/agent): not a pi extension - this is a skill, not a Tool extension. Pi consumes skills via ~/.agents/skills/ symlinks; see ~/.senpi/.pi/agent/skills/ for the convention.
Direct CLI use: python3 ~/.agents/skills/web-fetch/scripts/web_fetch.py .

Usage

# Default (markdown, auto-converted from HTML)
python3 scripts/web_fetch.py https://example.com

# Format options
python3 scripts/web_fetch.py https://example.com --format markdown
python3 scripts/web_fetch.py https://example.com --format text
python3 scripts/web_fetch.py https://example.com --format html
python3 scripts/web_fetch.py https://example.com --format raw

# Custom output dir, custom timeout
python3 scripts/web_fetch.py https://example.com --output-dir ./scratch --timeout 60

# Print only the content path (for piping)
python3 scripts/web_fetch.py https://example.com --print-path content

# Also write content to stdout
python3 scripts/web_fetch.py https://example.com --print-content

Stdout prints two paths b