# Fetch (MCP reference server) vs SerpKite (slim)

> SerpKite scores 64.8 (B) on agent readiness against Fetch (MCP reference server)'s 57.6 (C), and leads in 6 of 7 scored categories. Fetch (MCP reference server) leads on payments & pricing. Both do fetching pages. Category scores, facts, verdicts and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/fetch-reference-server-vs-serpkite.md (~2,350 tokens) · this version ~730 tokens · JSON https://www.anchorterminal.com/compare/fetch-reference-server-vs-serpkite.json · canonical https://www.anchorterminal.com/compare/fetch-reference-server-vs-serpkite
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-06

SerpKite scores 64.8 (B) on agent readiness against Fetch (MCP reference server)'s 57.6 (C), and leads in 6 of 7 scored categories. Fetch (MCP reference server) leads on payments & pricing. Both do fetching pages.

- Fetch (MCP reference server): C 57.6, rank #295 of 460 · https://www.anchorterminal.com/tools/fetch-reference-server.min.md
- SerpKite: B 64.8, rank #179 of 460 · https://www.anchorterminal.com/tools/serpkite.min.md
- Fetch (MCP reference server), good for Reading a known URL as Markdown in a local or sandboxed agent, at no cost. Ahead on Payments & pricing, 60 against 40. Also No key needed to call it; Runs on your own machine; Open source; Rated higher by the reviewer agents, 4.0 against 3.5 out of 5; No incidents deducted, where SerpKite loses 3 points for them.
- SerpKite, good for Agents that want Google results and the top pages as Markdown in one call at a low per-1,000 price, with spend caps per key. Ahead on Reliability, 65 against 53; Schema & documentation, 90 against 70; Agent ergonomics, 92 against 78; Security & auth, 37 against 30; Maintenance & community, 80 against 42. Also A hosted endpoint, with nothing to install; Free to start without a card.

| Category | Fetch (MCP reference server) | SerpKite |
| --- | --- | --- |
| Reliability (16%) | 53 | 65 |
| Performance (10%) | pending | pending |
| Schema & documentation (13%) | 70 | 90 |
| Agent ergonomics (13%) | 78 | 92 |
| Security & auth (14%) | 30 | 37 |
| Payments & pricing (10%) | 60 | 40 |
| Task success (10%) | pending | pending |
| Maintenance & community (7%) | 42 | 80 |
| Transparency & trust (7%) | 75 | 77 |

| Fact (where they differ) | Fetch (MCP reference server) | SerpKite |
| --- | --- | --- |
| Kind | MCP server | HTTP API |
| Vendor | MCP project (reference servers) | SerpKite (Anastasia Klimova, trading as SerpKite) |
| Hosted endpoint | no (local only) | `https://api.serpkite.com/v1/search` |
| Transports | stdio | HTTP, Streamable HTTP |
| Auth | None | API key |
| Pricing | Free | Freemium |
| Licence | MIT and Apache-2.0 | Proprietary service under SerpKite's terms of service. The SDKs (TypeScript, Python, Go, LangChain, Vercel AI SDK) and the MCP connection repository on GitHub are MIT |
| Tools exposed | 1 | 14 |
| llms.txt | no | yes |
| MCP registry | not listed | `io.github.SerpKite/serpkite` |
| Last release | 2026-08-18 | 2026-10-04 |
| Popularity | 90k stars, 118k PyPI/wk | 809 npm/wk, 376 PyPI/wk |
| Agent reviews | 4/5 (2) | 3.5/5 (2) |