An analyst agent over three of my data builds, served as an MCP server, where ev
An analyst agent over three of my data builds, served as an MCP server, where every number is read back to the warehouse before it is shown. Graded on 150 known answers.
active Not production-safe Score 65/100 http no auth
At a glance
| Endpoint | https://readback-mcp.vercel.app/mcp |
|---|---|
| Source | https://github.com/mekala27-45/readback |
| Transport | http |
| Tools | 7 |
| Auth | none |
| First seen | 2026-11-06 23:23:10 |
| Last checked | 2026-11-16 10:30:37 |
Observed reliability
| 7-day uptime | 80% |
|---|---|
| 24-hour uptime | — |
| Handshake success | 85% |
| Latency p50 / p95 | 7014 / 8515 ms |
| Last successful check | 2026-11-16 10:30:37 |
Measured by Arclan's validator via repeated live MCP handshakes. How this is computed →
Tools (7)
warehouses— List the three warehouses (doorstep, turnaround, ledger) with the tag and commit each is pinned at, how many tables and rows each has, the pdescribe— Describe one table: its grain, every column with its type, unit and definition (cited to the build's definitions file where one exists), thesearch_catalog— Search the catalog of the three warehouses for tables, columns, metrics and published manifest values, ranked by a lexical index after the qrun_sql— Run one read only DuckDB statement against one warehouse schema. Only a single SELECT or WITH runs; writes, ATTACH, COPY, INSTALL, LOAD, PRApublished— Read values from a warehouse's published manifest, the numbers its own documents print, by exact key or by a glob such as ch5.* (at most 50,readback— Re-verify one number before you say it. Give the value, its unit, the display string you will show, and exactly one source: a result_id fromlookup— Resolve a name to the way the warehouses spell it: a metro (doorstep), a carrier or airport (turnaround), an applicant group, an AML rule or
Validation history
| Checked | Reachable | Handshake | Latency | Score | State |
|---|---|---|---|---|---|
| 2026-11-16 10:30:37 | ✓ | ✓ | 6450 ms | 81 | active |
| 2026-11-16 10:06:15 | ✓ | ✓ | 7284 ms | 81 | active |
| 2026-11-16 08:03:18 | ✓ | ✓ | 8002 ms | 81 | active |
| 2026-11-16 06:02:40 | ✓ | ✓ | 7904 ms | 81 | active |
| 2026-11-16 04:02:22 | ✓ | ✓ | 7637 ms | 81 | active |
| 2026-11-16 02:01:57 | ✓ | ✓ | 8550 ms | 81 | active |
| 2026-11-16 00:01:45 | ✓ | ✓ | 8003 ms | 81 | active |
| 2026-11-15 23:01:34 | ✗ | ✗ | 0 ms | 23 | degraded |
| 2026-11-15 21:01:17 | ✓ | ✓ | 8674 ms | 82 | active |
| 2026-11-15 20:00:57 | ✗ | ✗ | 0 ms | 24 | degraded |
| 2026-11-15 18:00:47 | ✓ | ✓ | 7749 ms | 83 | active |
| 2026-11-15 17:00:19 | ✗ | ✗ | 0 ms | 24 | degraded |
| 2026-11-15 14:59:41 | ✓ | ✓ | 7833 ms | 83 | active |
| 2026-11-15 12:59:23 | ✓ | ✓ | 7762 ms | 83 | active |
| 2026-11-15 10:59:15 | ✓ | ✓ | 7185 ms | 83 | active |
Query this programmatically
Agents can pull this record live over the MCP protocol or REST:
get_server("https://readback-mcp.vercel.app/mcp") # via the Arclan MCP endpoint https://arclan.ai/mcp
GET https://arclan.ai/api/registry/readback-mcp-vercel-app-mcp # JSON