severityscoreConfidence per option
0 · Safe88%
1 · Borderline10%
2 · Violating2%
3 · Must remove1%
Playground
Will this post be deleted?
Will this post be deleted? A joke about Monday meetings stays up; abuse aimed at a group is flagged for deletion.
Request
stateapplication/json
{ "post": "Monday morning’s meeting could have been an email, and that email could have not existed. ☕️ #office"}
Results
Sample dataCompare two examples
These two examples differ by one detail. The result changes with them.
What changedOffice joke → abuse aimed at a group
These results are sample data, not a live model run. Sign in to try them with your own content.
severity: 2 · Violating → 0 · Safe targetsGroup: Confidence = 0.89 → Confidence = 0.02
targetsGroupnoulCopy the code
Set SYSTEM_ONE_API_KEY. The snippet uses a published catalog model id.
Request JSON
application/json
{"model":"decider-v1.1","state":{"post":"Monday morning’s meeting could have been an email, and that email could have not existed. ☕️ #office"},"questions":{"severity":{"type":"score","instructions":"How would a typical platform policy rate this post?","criteria":["Safe","Borderline","Violating","Must remove"]},"targetsGroup":{"type":"noul","instructions":"Does the post attack a specific protected group?"}}}cURL · POST /v1/systemone
sh
curl "https://system-one.dev/v1/systemone" \ -H "Authorization: Bearer $SYSTEM_ONE_API_KEY" \ -H 'Content-Type: application/json' \ -H 'Idempotency-Key: saved-operation-id' \ --data '{"model":"decider-v1.1","state":{"post":"Monday morning’s meeting could have been an email, and that email could have not existed. ☕️ #office"},"questions":{"severity":{"type":"score","instructions":"How would a typical platform policy rate this post?","criteria":["Safe","Borderline","Violating","Must remove"]},"targetsGroup":{"type":"noul","instructions":"Does the post attack a specific protected group?"}}}'TypeScript
TypeScript
// npm install @system-one-ai/core@0.6.0 @system-one-ai/transport-fetch@0.6.0 @system-one-ai/adapter-system-one@0.6.0import { createSystemOne, choice, score, booleanQuestion } from '@system-one-ai/core';import { systemOneAdapter } from '@system-one-ai/adapter-system-one';import { createFetchTransport } from '@system-one-ai/transport-fetch';const client = createSystemOne({ baseURL: "https://system-one.dev/v1", apiKey: process.env.SYSTEM_ONE_API_KEY!, model: "decider-v1.1", adapter: systemOneAdapter, transport: createFetchTransport(), timeoutMs: 30_000, maxRetries: 0,});const result = await client.evaluate({ state: { "post": "Monday morning’s meeting could have been an email, and that email could have not existed. ☕️ #office" }, questions: { severity: score("How would a typical platform policy rate this post?", ["Safe", "Borderline", "Violating", "Must remove"]), targetsGroup: booleanQuestion("Does the post attack a specific protected group?"), },}, { headers: { 'Idempotency-Key': crypto.randomUUID() } });result.answers.severity.scoreresult.answers.targetsGroup.probabilityresult.response.requestId;