Open-WebSearch vs Fetch (reference)
Both are web search & scraping MCP servers, and we ran both through the same test (page read, web search) in each app. Open-WebSearch passed 4 of 4 apps we tested it in; Fetch (reference) passed 3 of 4.
Test results side by side
| Open-WebSearch | Fetch (reference) | |
|---|---|---|
| Claude Code | Passed 2026-10-21 | Passed 2026-10-19 |
| Cursor | Passed 2026-10-21 | Passed 2026-10-19 |
| Claude Desktop | Passed 2026-10-21 | Passed 2026-10-14 |
| ChatGPT (Codex) | Passed 2026-10-21 | Failed 2026-10-20 |
| Tools exposed (protocol check) | 6 tools | 1 tools |
| GitHub stars | 1,862 | 91,266 (shared repo) |
| License | Apache-2.0 | NOASSERTION |
| Last push | 2026-10-08 | 2026-10-13 |
Results from our own harness runs, dated per cell. Stars and last push from GitHub at our last data refresh. We don’t pick a winner here: which one fits depends on your app and setup.
What each does
Open-WebSearch: Multi-engine web search and page fetching for agents with no API keys; also ships a CLI and a local daemon.
Fetch (reference): Fetches a web page and returns it as Markdown so the agent can read it.
