NOW · SOURCES
Sources
12 sources. One unified corpus. Every NEAR knowledge source you would search separately, indexed continuously and exposed through 27 MCP tools. The numbers below refresh every minute, straight from the production database.
12sources
synced 7m agoall sources
Knowledge layer
12 SOURCES · SYNCING
27MCP TOOLS-AVG LATENCY● SERVER
TOTAL INDEXED RECORDS
12 of 12 sources · full-text search
12 of 12 sources · full-text search
11,878records
Source freshness12
| Source | Records | Cadence | Last sync | Status |
|---|---|---|---|---|
| GitHub items | 3,186 | every 6 h | synced | |
| Forum | 2,836 | every 15 min | synced | |
| Videos | 1,837 | every 24 h | synced | |
| Releases | 1,328 | every 15 min | synced | |
| GitHub repositories | 808 | every 24 h | synced | |
| Blog | 670 | every 24 h | synced | |
| Docs | 483 | every 6 h | synced | |
| Ecosystem | 340 | every 24 h | synced | |
| Discussions | 255 | every 6 h | synced | |
| npm + crates.io packages | 90 | every 6 h | synced | |
| House of Stake proposals | 30 | every 15 min | synced | |
| Security incidents | 15 | every 24 h | synced |
Showing 12 of 1226,453 embedded chunks
How the data lands here
Each source is polled on its own cadence, listed in the table above. Anything changed since the last sync is upserted and re-embedded.
Why one corpus matters
Cross-source questions (“which release removed this method, and is anyone on the forum hitting it?”) only work when docs, releases, GitHub, the forum and proposals share a single semantic index.
Verify the numbers
The widget calls /api/v1/sources. Same endpoint anyone can hit - no caching, no embellishment.