Persistent memory, reflection & safety for AI agents. No vector DB, no embedding
Persistent memory, reflection & safety for AI agents. No vector DB, no embedding server — BM25 + multi-hop retrieval runs in-process. Hosted MCP server with 31 tools. MIT licensed.
auth_gated Not production-safe Score 36/100 unknown auth required
At a glance
| Endpoint | https://mcp.cognicoree.tech/mcp/stream |
|---|---|
| Source | https://github.com/cognicore-dev/cognicore-env |
| Transport | unknown |
| Tools | 0 |
| Auth | bearer (bearer) |
| First seen | 2026-11-03 01:03:38 |
| Last checked | 2026-11-10 01:04:37 |
Observed reliability
| 7-day uptime | — |
|---|---|
| 24-hour uptime | — |
| Handshake success | 0% |
| Latency p50 / p95 | 0 / 0 ms |
| Last successful check | — |
Measured by Arclan's validator via repeated live MCP handshakes. How this is computed →
Validation history
| Checked | Reachable | Handshake | Latency | Score | State |
|---|---|---|---|---|---|
| 2026-11-10 01:04:37 | ✓ | ✗ | 0 ms | 35 | auth_gated |
| 2026-11-03 01:04:04 | ✓ | ✗ | 0 ms | 35 | auth_gated |
Query this programmatically
Agents can pull this record live over the MCP protocol or REST:
get_server("https://mcp.cognicoree.tech/mcp/stream") # via the Arclan MCP endpoint https://arclan.ai/mcp
GET https://arclan.ai/api/registry/mcp-cognicoree-tech-mcp-stream # JSON