context-mode vs Everything (reference)
Both are memory & reasoning MCP servers, and we ran both through the same test (recall across sessions) in each app. context-mode passed 4 of 4 apps we tested it in; Everything (reference) passed 2 of 2.
Test results side by side
| context-mode | Everything (reference) | |
|---|---|---|
| Claude Code | Passed 2026-10-13 | Passed 2026-10-13 |
| Cursor | Passed 2026-10-13 | Passed 2026-10-13 |
| Claude Desktop | Passed 2026-10-14 | Not tested yet |
| ChatGPT (Codex) | Passed 2026-10-14 | Not tested yet |
| Tools exposed (protocol check) | 11 tools | 13 tools |
| GitHub stars | 29,079 | 91,266 (shared repo) |
| License | NOASSERTION | NOASSERTION |
| Last push | 2026-10-14 | 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
context-mode: Keeps large tool outputs out of the context window by summarizing/indexing them.
Everything (reference): Reference test server from the MCP maintainers that exercises every protocol feature (tools, prompts, resources). Useful for checking a client, not for daily work.
