Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/nimblebraininc/mcp-webfetch/mcp_webfetchnpx skills add NimbleBrainInc/mcp-webfetch --skill mcp_webfetchgit clone --depth 1 https://github.com/NimbleBrainInc/mcp-webfetchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nimblebraininc/mcp-webfetch/mcp_webfetch)<a href="https://agentmods.dev/skills/nimblebraininc/mcp-webfetch/mcp_webfetch"><img src="https://agentmods.dev/badge/skills/nimblebraininc/mcp-webfetch/mcp_webfetch.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00449 |
| Opus 5 | $0.00000 | $0.00225 |
| Sonnet 5 | $0.00000 | $0.00090 |
| Haiku 4.5 | $0.00000 | $0.00045 |
Grade A, and why
mcp_webfetch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
WebFetch
When to Use (and When Not To)
| Situation | Tool |
|---|---|
| Have a specific URL, want to understand its content | web_fetch |
| Need to find URLs on a topic | Use a web search tool first, then web_fetch on results |
| Want raw HTML or markdown | Don't use this, it returns an LLM summary |
| URL points to PDF, image, JSON API | Don't use this, it only handles HTML pages |
Writing Effective Prompts
The prompt is sent to an LLM along with the page's markdown. Specific prompts get better results.
| Weak | Strong |
|---|---|
| "Summarize this page" | "Extract the pricing tiers and format as a table" |
| "What is this?" | "What problem does this product solve and who is the target audience?" |
| "Tell me about it" | "List the API endpoints documented on this page with their HTTP methods" |
For large pages, say what section you care about: "Focus on the installation instructions" or "Only look at the FAQ section."
URL Handling
- No scheme? Auto-prepended with
https://("example.com" becomes "https://example.com") - HTTP? Auto-upgraded to HTTPS
- Cross-host redirect? Returns an error with the redirect URL. Call
web_fetchagain with that new URL.
Caching
Fetched pages are cached for 15 minutes. Multiple questions about the same URL are cheap because only the LLM step re-runs, not the HTTP fetch. Use this to your advantage:
web_fetch(url="https://example.com/docs", prompt="What authentication methods are supported?")
web_fetch(url="https://example.com/docs", prompt="What are the rate limits?") ← cache hit, no re-fetch
Limitations
- HTML only. Non-HTML content types (PDF, JSON, images) return an error.
- Pages requiring JavaScript rendering may return incomplete content.
- Content over 180K characters is truncated. Be specific in your prompt for large pages.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 44 lines · 0 tokens per session scan A c068cbb469b6
mcp_webfetch is a skill published in the GitHub repository NimbleBrainInc/mcp-webfetch (0 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 449 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
mcp_finnhub
Follow-up questions ("what's the market cap?", "what do analysts say?") should use data from earlier tool responses before making new calls. If you already fetched a company profile or financials, surface that data directly.
mcp_ipinfo
Follow-up questions ("this IP", "that one", "is it a VPN?") → reuse IP from prior result.
mcp_granola
All tools that accept meeting identifiers use meetingid (not docid, id, or documentid). The meetingid value comes from the id field in search results and meeting lists.
use-agent-browser-for-airi
Test AIRI display-model imports with agent-browser across stage-tamagotchi Electron, stage-web, and stage-pocket mobile web layouts. Use when uploading and verifying contributor-supplied Live2D ZIP, VRM, or MMD ZIP/PMX/PMD files through AIRI's model selector, including onboarding bypass, format-specific import…
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
interactive-login
How to complete browser/interactive logins (aws / gh / glab / gcloud). The platform backgrounds the login poller so it survives the human's browser round-trip — and when that does NOT work.