# Bluetouff 13flow MCP server

Source: https://glimind.com/server/mcp-registry%2Fio.github.bluetouff%2F13flow
Registry id: `mcp-registry/io.github.bluetouff/13flow`
Status: live (checked 2026-10-27T04:11:10.126Z)

## Verdict

Bluetouff 13flow MCP server answered 99.81% of Glimind's liveness checks over 30 days, better than 24% of MCP servers, with 876 ms median handshake and 5 outages in 90 days. Its agent-readiness grade is A (90/100) across 13 tools, costing an agent about 1,694 tokens to load. Checked every few minutes; no tool is ever called.

## Measurements (30 days)

| Metric | Value | Better than |
|---|---|---|
| Uptime, 30 days | 99.81% | 24% of MCP servers |
| Median handshake latency | 876 ms | 24% of MCP servers |
| p95 handshake latency | 3439 ms | — |
| Checks in 30 days | 4,891 | — |
| Outages, 90 days | 5 | — |
| Median recovery | 6 min | — |
| Agent-readiness score | 90/100 (A) | 69% of MCP servers |
| Tools | 13 | — |
| Tool-list size for an agent | ~1,694 tokens | 56% of MCP servers |
| Days observed | 29 | — |

## Install

```json
{
  "mcpServers": {
    "bluetouff-13flow-mcp-server": {
      "type": "http",
      "url": "https://13flow.eu/api/mcp"
    }
  }
}
```

Claude Code:

```sh
claude mcp add --transport http bluetouff-13flow-mcp-server https://13flow.eu/api/mcp
```

## Tools (13)

- `discover_watchlist`: Rank trusted 13F watchlist candidates with explicit filters. Scores are ordinal research screens, not forecasts.
- `get_agent_stats`: Return aggregate 7/30-day MCP usage, fixed client families, tool mix and reliability. Initializations are not unique users; no IP, user-agent, version, argument
- `get_confluence_methodology`: Return the frozen Confluence v1 methodology contract, including proof boundary and validation requirements.
- `get_confluence_signals`: Return public cached Confluence v1 signals. Scores are ordinal heuristic ranks, not probabilities. Diagnostics include up to 10 examples per category, with avai
- `get_data_quality`: Return public read-only data-quality warnings. These are review signals, never automatic corrections.
- `get_fund`: Get a bounded public fund portfolio by SEC CIK, including filing metadata, positions, moves and history counts.
- `get_live_status`: Return verifiable public state: LIVE/DEMO/DEGRADED, commit, generated_at, data_as_of, 13F period and coverage.
- `get_product_status`: Return go-to-market readiness, sellable boundaries, validation status and blocked full-quant artifact.
- `get_research_readiness`: Return the current research, validation, data-quality and operator-review boundary without commercial claims.
- `get_signal_history`: Read append-only Confluence signal revisions for audit and replay.
- `get_stock`: Get bounded current 13F holders, quarter moves, quality warnings and ordinal research score for one ticker.
- `list_funds`: Search and page through compact tracked 13F manager summaries. Use get_fund for positions and moves.
- `preview_watchlist`: Evaluate 1 to 25 tickers against trusted 13F flow, quality gates and explainable watch/monitor/block triggers.

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.