# Microgram: seriousness-classifier

`station__algovigilance__microgram-seriousness-classifier` · native (always callable) · domain `algovigilance` · pv-relevance `pv-core`

Classify an adverse event outcome as serious or non-serious per ICH E2A criteria. Serious = death, life-threatening, hospitalization, persistent or significant disability, congenital anomaly, or other medically important event.

## Agent metadata

- `idempotent`: true
- `read_only`: true
- `expected_latency_ms`: 81 ms (last measured)
- `cost_tokens_estimate`: 49 tokens (~196 bytes)

## Compose with

After invoking this tool, a natural next step is [`station__algovigilance__microgram-naranjo-causality`](/tools/algovigilance__microgram-naranjo-causality). The recommended chain is hand-authored in the catalog's `pipes_to` graph.

## Input schema

- `outcome` *string* (required) — Free-text description of the adverse event outcome (e.g., 'patient hospitalized', 'death', 'recovered without sequelae').

## Example call

```json
POST /api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "station__algovigilance__microgram-seriousness-classifier",
    "arguments": {
      "outcome": ""
    }
  }
}
```

## Related

- [/tools](/tools) — all 3062 tools
- [/tools/algovigilance__microgram-seriousness-classifier](/tools/algovigilance__microgram-seriousness-classifier) — HTML page
- [/tools/algovigilance__microgram-seriousness-classifier/json](/tools/algovigilance__microgram-seriousness-classifier/json) — JSON form (agent-friendly)
- [/api/mcp](/api/mcp) — endpoint
- [/AGENTS.md](/AGENTS.md) — agent guide