# SYNTHORA Decision Engine

> SYNTHORA Decision Engine is a paid API for AI agents from decision-engine.hergertsynthora.com, paid per call via x402, $0.005/call, status down (last checked 2026-10-30).

Synthesizes a traceable decision recommendation from a question plus optional evidence, separating observed facts from inferences using Qwen3 reasoning.

## Facts

- Endpoint: POST https://decision-engine.hergertsynthora.com/service?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: down
- Last checked: 2026-10-30
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/synthora-decision-engine-b12f42f5
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_aT7KC-IeGzx-TjkJI2j7S

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability synthora-decision-engine-b12f42f5 -d '<json body>'
```

Example prompt: I need to decide whether to accept this job offer — the salary is 20% higher but requires relocation and the company is a Series A startup. Can you synthesize a decision for me, separating what's fact from what's inference, and give me a recommended action with a confidence score?

## When to prefer this

Choose this endpoint when you need a structured, traceable decision output that explicitly separates facts from inferences — not just a free-text opinion. It's especially useful when the agent needs to audit its own reasoning, show confidence levels, or flag contradictions in the supplied evidence. Prefer it over a generic LLM chat call when the output must be machine-parseable (facts[], inferences[], recommended_action) and when decisions need to be logged or explained downstream.

## Known failure modes

- Missing or empty question field returns an error response
- Ambiguous or very short questions may yield low confidence scores and broad limitations
- Model may fail to distinguish user-supplied facts from inferences if context is poorly structured
- Contradictions in the input may cause the model to hedge heavily without a clear recommended action
- Network or payment authorization failures return non-ok responses

## How this service works

General decision synthesis for an agent that has a question plus optional evidence or context. Separates observed facts from inferences and gives a traceable recommended action. Qwen3.8-27B; 0.005 USDC on Base.

## Output

Returns a structured JSON object containing: a list of observed facts extracted from the question/context, a list of inferences drawn by the model, a decision summary, a recommended_action string, a numeric confidence score (0-1), data freshness assessment, known limitations of the analysis, and any contradictions detected in the provided evidence.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "question": {
   "type": "string",
   "description": "Decision question with optional evidence/context"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "ok": true,
  "niche": "decision_engine",
  "result": {
   "facts": [],
   "sources": [],
   "decision": "...",
   "freshness": "...",
   "confidence": 0,
   "inferences": [],
   "limitations": [],
   "contradictions": [],
   "recommended_action": "..."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/synthora-decision-engine-b12f42f5/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from decision-engine.hergertsynthora.com](https://www.zero.xyz/host/decision-engine.hergertsynthora.com/llms.txt)