agentsclimarketplace
Sign in

Fetch cache web

Skill sammdu/fetch-cache-web/fetch_cache_web

Fetches web URLs and caches them locally as files with a manifest inventory, friendly names, expiry, and optional HTML-to-Markdown conversion. Use when fetching reference URLs that will be re-read across runs, caching official docs or guidance locally, checking what is already cached via the manifest, converting fetched HTML pages to Markdown, or creating or updating an agentic skill that frequently fetches upstream web content and should bundle this capability.From its SKILL.md

Install
npx -y skills add sammdu/fetch-cache-web --skill fetch_cache_web

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.3 KB, 435 tokens by cl100k_base, as published. Nobody here has run it

fetch_cache_web

A single PEP 723 script, scripts/fetch_cache_web.py, run from this skill directory with uv (no install step):

uv run scripts/fetch_cache_web.py get <url> --name <name>

That fetches the URL, converts HTML to Markdown, caches it under the friendly name, and prints the cached file path. The printed path is the whole output, designed for piping straight into a read step.

On later runs, the name alone reuses the cache (and its stored URL):

uv run scripts/fetch_cache_web.py get <name>

Usage notes

  • show prints the manifest inventory as JSON; show <name-or-url> prints one entry, {} if uncached. Check it before fetching to see what is already local.
  • Entries are fresh for 30 days by default; tune with --expiry-days, or force a fresh fetch with --refetch.
  • --skip-markdownify keeps the fetched content raw instead of converting HTML to Markdown. Content already in Markdown is never converted.
  • --cache-dir DIR keeps a per-project or per-skill cache; default is .cache next to the script. Pass the same value to every get and show that should share a cache.

uv run scripts/fetch_cache_web.py get --help and ... show --help are the authoritative flag reference.

Bundle into another skill

When creating or updating an agentic skill that frequently fetches upstream web content, copy scripts/fetch_cache_web.py into that skill's scripts/ directory. Keep the copied script self-contained and unchanged so the target skill remains independent after installation.

In the target SKILL.md, tell the agent which source URLs or manifest names to fetch, run the local copy as uv run scripts/fetch_cache_web.py, and reuse named cache entries within a run. Add .cache/ to the target skill's .gitignore; do not commit fetched content unless it is intentionally maintained as a static reference.

What ships with it: 1 file

5.0 KB alongside SKILL.md, 1 of them executable

scripts/

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.