Read back one exhaustive search

Reads back one search run: every shape tried over this organisation's own history, best first, and the one that fit.

GET /v1/risk/search/{id}

Risk → · All API references → · Six flows, four surfaces →


Address	`https://api.hanzo.ai/v1/risk/search/{id}`
Method	`GET`
Operation	`riskSearchResult`
Auth	`Authorization: Bearer $HANZO_API_KEY`

Reads back one search run: every shape tried over this organisation's own history, best first, and the one that fit.

A run another organisation started is simply not there — the same 404 an unknown id gives, so the read is not a probe oracle.

Request

1 field.

Field	In	Type	Required	Description
`id`	path	string	yes	ID is the run, taken from the path.

Response

Status	Body	Meaning
`200`	riskSearchReport	ok

200 body — 48 fields.

Field	In	Type	Always	Description
`done`	body	boolean	—	Done is false while the run is still going; the trials below are then the ones finished so far.
`ended`	body	string	—	Ended is when it finished, RFC 3339.
`events`	body	integer	—	Events is how much of this organisation's history was replayed.
`fitted`	body	riskModelValue	—
`fitted.address`	body	string	—	Address names this value by its own content: the model's shape, the geometry seed, its position in the window, its threshold, its masses as IEEE-754 bits and…
`fitted.at`	body	string	—	At is when it was published, RFC 3339, on the server clock.
`fitted.learned`	body	integer	—	Learned is how many events are behind the masses.
`fitted.sequence`	body	integer	—	Sequence is this value's place in YOUR organisation's own history, from 1 and contiguous until retention disposes of the oldest.
`fitted.shape`	body	string	—	Shape NAMES the model space the masses are only meaningful against, as `<family>:<digest>` — the KIND of model, and that family's own digest over the feature…
`fitted.warmed`	body	string	—	Warmed is how far your own event surface had been folded in when this value was published, RFC 3339.
`gap`	body	string	—	Gap says why the winning shape could not be fitted into an adoptable value, when it could not.
`id`	body	string	—	ID is the run.
`refusal`	body	string	—	Refusal says why the run proves nothing, when it does.
`started`	body	string	—	Started is when the run was accepted, RFC 3339.
`trials`	body	riskTrial[]	—	Trials is every shape tried, best first.
`trials[].alerted`	body	integer	—	Alerted is how many of those it would have raised.
`trials[].curve`	body	number[]	—	Curve is the realised alert rate over successive tenths of the history — the learning curve, which says whether the shape settled or is still moving.
`trials[].fit`	body	number	—	Fit ranks the shape, smaller being better: the relative miss of the stated appetite, plus flat penalties for never warming and for saturating, plus the share…
`trials[].learned`	body	integer	—	Learned is how many events the shape learned from during the replay.
`trials[].realised`	body	number	—	Realised is what that appetite actually produced.
`trials[].saturated`	body	boolean	—	Saturated is whether the appetite could not be honoured by any threshold, which is a shape that alerts on nothing and reads like a quiet one.
`trials[].scored`	body	integer	—	Scored is how many it was able to score.
`trials[].stated`	body	number	—	Stated is the appetite the shape was tried at.
`trials[].topology`	body	riskTopology	—
`trials[].topology.blend`	body	number	—	Blend is how much of a closing window folds into the reference: 1 replaces it outright, less makes the reference expensive to move.
`trials[].topology.depth`	body	integer	—	Depth is how deep each tree is.
`trials[].topology.family`	body	string	—	Family is the KIND of model this candidate is: `halfspace` is an ensemble of half-space trees whose masses are counters, and it is the family this search grid…
`trials[].topology.review`	body	number	—	Review is the appetite this shape was tried at.
`trials[].topology.trees`	body	integer	—	Trees is how many half-space trees the ensemble holds.
`trials[].topology.window`	body	integer	—	Window is how many events make one reference window.
`trials[].warm`	body	boolean	—	Warm is whether the shape learned enough to have an opinion at all over this organisation's whole history.
`winner`	body	riskTrial	—
`winner.alerted`	body	integer	—	Alerted is how many of those it would have raised.
`winner.curve`	body	number[]	—	Curve is the realised alert rate over successive tenths of the history — the learning curve, which says whether the shape settled or is still moving.
`winner.fit`	body	number	—	Fit ranks the shape, smaller being better: the relative miss of the stated appetite, plus flat penalties for never warming and for saturating, plus the share…
`winner.learned`	body	integer	—	Learned is how many events the shape learned from during the replay.
`winner.realised`	body	number	—	Realised is what that appetite actually produced.
`winner.saturated`	body	boolean	—	Saturated is whether the appetite could not be honoured by any threshold, which is a shape that alerts on nothing and reads like a quiet one.
`winner.scored`	body	integer	—	Scored is how many it was able to score.
`winner.stated`	body	number	—	Stated is the appetite the shape was tried at.
`winner.topology`	body	riskTopology	—
`winner.topology.blend`	body	number	—	Blend is how much of a closing window folds into the reference: 1 replaces it outright, less makes the reference expensive to move.
`winner.topology.depth`	body	integer	—	Depth is how deep each tree is.
`winner.topology.family`	body	string	—	Family is the KIND of model this candidate is: `halfspace` is an ensemble of half-space trees whose masses are counters, and it is the family this search grid…
`winner.topology.review`	body	number	—	Review is the appetite this shape was tried at.
`winner.topology.trees`	body	integer	—	Trees is how many half-space trees the ensemble holds.
`winner.topology.window`	body	integer	—	Window is how many events make one reference window.
`winner.warm`	body	boolean	—	Warm is whether the shape learned enough to have an opinion at all over this organisation's whole history.

Failure carries the platform error shape — see Errors.

Examples

hanzo risk search get <id>

import { Configuration, RiskApi } from 'hanzoai';

const api = new RiskApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.riskSearchResult({ id: 'id' });

from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import RiskApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = RiskApi(client).risk_search_result(id='id')

cfg := cloud.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := cloud.NewAPIClient(cfg)

resp, _, err := client.RiskAPI.RiskSearchResult(context.Background()).Execute()
if err != nil {
	return err
}

use hanzo_cloud::apis::{configuration::Configuration, risk_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = risk_api::risk_search_result(&cfg, Default::default()).await?;

import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.RiskApi;

ApiClient client = new ApiClient();
client.setRequestInterceptor(b -> b.header("Authorization", "Bearer " + System.getenv("HANZO_API_KEY")));

var result = new RiskApi(client).riskSearchResult();

curl https://api.hanzo.ai/v1/risk/search/<id> \
  -H "Authorization: Bearer $HANZO_API_KEY"

Tool risk, op riskSearchResult — POST the JSON-RPC envelope to https://api.hanzo.ai/v1/mcp.

curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "risk",
         "arguments": {
           "op": "riskSearchResult",
           "input": {
             "id": "<id>"
           }
         }
       }
     }'

Risk API · All Hanzo APIs · Interactive reference

Read back one exhaustive search

Request

Response

Examples

On this page