Web Fetch
Use when the user wants to fetch and read a webpage's content as markdown. Includes a 24h cache, optional JSON output, an SPA fallback via Playwright, and an SSRF guard that refuses non-public addresses.
Install
$ agentstack add skill-samir-g-agent-skills-web-fetch ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Web fetch
Run the bundled script with uv run so dependencies install on first use:
uv run "${CLAUDEPLUGINROOT}/skills/web-fetch/web_fetch.py"
Common flags
--json/-j— emit a JSON object instead of markdown.--no-cache— bypass the 24h cache.--render— force Playwright rendering (skip the SPA heuristic).--timeout N— request timeout in seconds (default 30).
Behaviour
- Default output: markdown with title, status, and the page's main content.
- Cache lives at
~/.cache/web_fetch/and expires after 24 hours. - Responses larger than 10 MiB are rejected.
- URLs that resolve to private/internal addresses (loopback, RFC1918, link-local, cloud metadata) are refused — including across redirects.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: samir-g
- Source: samir-g/agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.