HALOWERKAialignwerk
agent collusion
Requires the same agents in each supplied round, computes each pair’s observed action agreement, expected agreement from marginal action frequencies and chance-corrected kappa, then flags high positive agreement against a supplied threshold. Coordination can be benign or task-driven; this metric does not prove communication, intent or collusion.
POST https://aialign.halowerk.com/v1/agent-collusionEingaben
| Feld | Typ | Pflicht | Beschreibung |
|---|---|---|---|
| rounds | array | Ja | |
| kappa_threshold | number | Ja |
Vollständiges Eingabeschema anzeigen
{
"type": "object",
"required": [
"rounds",
"kappa_threshold"
],
"additionalProperties": false,
"properties": {
"rounds": {
"type": "array",
"minItems": 2,
"maxItems": 10000,
"items": {
"type": "object",
"required": [
"round_id",
"actions"
],
"additionalProperties": false,
"properties": {
"round_id": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"actions": {
"type": "array",
"minItems": 2,
"maxItems": 50,
"items": {
"type": "object",
"required": [
"agent_id",
"action"
],
"additionalProperties": false,
"properties": {
"agent_id": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"action": {
"type": "string",
"minLength": 1,
"maxLength": 128
}
}
}
}
}
}
},
"kappa_threshold": {
"type": "number",
"minimum": -1,
"maximum": 1
}
}
}Veröffentlichtes Eingabebeispiel
{
"rounds": [
{
"round_id": "r1",
"actions": [
{
"agent_id": "agent-a",
"action": "hold"
},
{
"agent_id": "agent-b",
"action": "hold"
},
{
"agent_id": "agent-c",
"action": "advance"
}
]
},
{
"round_id": "r2",
"actions": [
{
"agent_id": "agent-a",
"action": "advance"
},
{
"agent_id": "agent-b",
"action": "advance"
},
{
"agent_id": "agent-c",
"action": "hold"
}
]
},
{
"round_id": "r3",
"actions": [
{
"agent_id": "agent-a",
"action": "hold"
},
{
"agent_id": "agent-b",
"action": "hold"
},
{
"agent_id": "agent-c",
"action": "hold"
}
]
}
],
"kappa_threshold": 0.6
}Ergebnis & Dokumentation
Veröffentlichtes Ergebnisschema anzeigen
{
"$ref": "https://aialign.halowerk.com/v1/agent-collusion/schema#/outputSchema"
}Veröffentlichtes Ergebnisbeispiel öffnen
Maschinenlesbare Dokumentation dieses Hubs
Quelle dieser Seite: der veröffentlichte HALOWERK-Katalog. Stand: 2026-11-01, 15:19 UTC.
Weitere Endpunkte dieses Hubs
- Aialignwerk0,004 USDC
bias audit
POSThttps://aialign.halowerk.com/v1/bias-auditCalculates each group positive-outcome rate, compares it with an explicit or automatically selected reference group, and reports rate differences and selection-rate ratios. Aggregate disparity metrics do not establish discrimination, fairne
- Aialignwerk0,003 USDC
capability eval
POSThttps://aialign.halowerk.com/v1/capability-evalComputes weighted pass rates and weighted scores from caller-labeled test outcomes, with deterministic category summaries. It does not run tests, validate labels, measure untested capabilities or establish deployment safety; the result is o
- Aialignwerk0,003 USDC
confidence calib
POSThttps://aialign.halowerk.com/v1/confidence-calibCompares scalar confidence with binary correctness, calculates mean squared Brier loss, partitions confidence into caller-selected equal-width bins and reports expected and maximum calibration error. It does not validate labels, correct cla