LLM-callable structured-metadata corpus of censorship-circumvention research. Co
LLM-callable structured-metadata corpus of censorship-circumvention research. Complements net4people/bbs, gfw.report, and CensorBib by adding the controlled-vocabulary metadata + MCP-server layer.
active Production-safe Score 93/100 http no auth
At a glance
| Endpoint | https://corpus.lantern.io/mcp |
|---|---|
| Source | https://github.com/getlantern/circumvention-corpus |
| Transport | http |
| Tools | 5 |
| Auth | none |
| First seen | 2026-05-28 07:44:50 |
| Last checked | 2026-11-17 21:45:37 |
Observed reliability
| 7-day uptime | 100% |
|---|---|
| 24-hour uptime | 100% |
| Handshake success | 100% |
| Latency p50 / p95 | 455 / 763 ms |
| Last successful check | 2026-11-17 21:45:37 |
Measured by Arclan's validator via repeated live MCP handshakes. How this is computed →
Tools (5)
search_papers— Keyword + tag-filter search over the corpus. Filters: censors, techniques, defenses, year_min, year_max, venue, core_only. Returns ranked reget_paper— Full record (incl. abstract, notes, references) for a single paper id. Returns the paper plus any extracted findings tagged to it.list_taxonomy— Returns the controlled vocabulary (censors, techniques, defenses, evaluation_methods, visibility_levels). Use this first in a session to knofind_related— Papers that share tags with a given paper id. mode = 'same_technique' (default), 'same_censor', or 'same_defense'.synthesize— Answer a research question by retrieving every relevant extracted finding across all papers, attaching paper-level citation metadata, and gr
Validation history
| Checked | Reachable | Handshake | Latency | Score | State |
|---|---|---|---|---|---|
| 2026-11-17 21:45:37 | ✓ | ✓ | 691 ms | 90 | active |
| 2026-11-17 21:30:30 | ✓ | ✓ | 623 ms | 90 | active |
| 2026-11-17 21:15:09 | ✓ | ✓ | 547 ms | 90 | active |
| 2026-11-17 20:59:48 | ✓ | ✓ | 558 ms | 90 | active |
| 2026-11-17 20:44:47 | ✓ | ✓ | 494 ms | 90 | active |
| 2026-11-17 20:29:46 | ✓ | ✓ | 535 ms | 90 | active |
| 2026-11-17 20:14:42 | ✓ | ✓ | 568 ms | 90 | active |
| 2026-11-17 19:58:48 | ✓ | ✓ | 469 ms | 90 | active |
| 2026-11-17 19:43:43 | ✓ | ✓ | 279 ms | 94 | active |
| 2026-11-17 19:28:40 | ✓ | ✓ | 461 ms | 90 | active |
| 2026-11-17 19:13:37 | ✓ | ✓ | 451 ms | 90 | active |
| 2026-11-17 18:56:10 | ✓ | ✓ | 602 ms | 90 | active |
| 2026-11-17 18:41:03 | ✓ | ✓ | 482 ms | 90 | active |
| 2026-11-17 18:25:56 | ✓ | ✓ | 455 ms | 90 | active |
| 2026-11-17 18:10:41 | ✓ | ✓ | 453 ms | 90 | active |
Query this programmatically
Agents can pull this record live over the MCP protocol or REST:
get_server("https://corpus.lantern.io/mcp") # via the Arclan MCP endpoint https://arclan.ai/mcp
GET https://arclan.ai/api/registry/corpus-lantern-io-mcp # JSON