Integrate ACI Framework scores into your platform.

Authentication

All API requests require an API key passed in the x-api-key request header. API keys are available in your Aethon Terminal workspace settings.

x-api-key: sk_live_your_api_key_here

Base URL

https://aethoncredit.com/api/v1

TypeScript / JavaScript SDK

Official SDK for Node.js and browser environments. Typed with full TypeScript support. Zero runtime dependencies.

npm install @aethoncredit/sdk

import { ACIClient } from '@aethoncredit/sdk'

const client = new ACIClient({ apiKey: 'sk_live_...' })

// List all provider scores
const scores = await client.getScores()

// Get a specific provider score
const score = await client.getScore('provider-id')

// Verify a webhook signature
import { WebhookVerifier } from '@aethoncredit/sdk'
WebhookVerifier.verify(rawBody, signature, 'whsec_...')

Python SDK

Official SDK for Python 3.8+. Sync and async clients available.

pip install aethoncredit-sdk

from ace import AciClient

client = AciClient(api_key="sk_live_...")

# List all provider scores
scores = client.list_scores()

# Get a specific provider score
score = client.get_score("provider-id")

Endpoints

Agent Workflows

Copy-paste prompt templates for any LLM. Each template directs the model to read deterministic fields off an ACI score response (especially the aci_decision_object block) and stay within them. No API calls embedded — pure prompt templates. Machine-readable ACI metadata at aethoncredit.com/llms.txt.

You are an institutional credit analyst. I'm pasting an ACI Risk Indicator response below. Read the aci_decision_object field in particular: it contains deterministic ceiling and floor factors with point impacts, hard caps fired, and band-gap distance.

Produce a 4-sentence risk summary covering:
1. The headline score and band.
2. The dominant ceiling factor and why it drags.
3. The dominant floor factor and what it anchors.
4. Any hard caps applied (or state none).

Do not interpolate beyond what the JSON states. Do not propose allocation actions.

[paste ACI score JSON here]

You are drafting an Investment Committee memorandum from an ACI portfolio score response. Render a structured memo with these sections:

- Executive Summary (3 sentences)
- Risk Profile (cite the aci_decision_object.summary verbatim)
- Key Drivers (bullet ceiling_factors with criterion_label and point_impact)
- Anchors (bullet floor_factors)
- Hard Caps (list hard_caps_fired or state "None applied")
- Band Gap (state points_to_next_band → next_band, or "currently in lowest band")

Cite the Verification ID at the bottom in the format "Verification ID: {id}" and the URL "Verify at: aethoncredit.com/verify/{id}".

Stay deterministic — do not interpolate beyond the JSON. Do not propose allocation sizing.

[paste ACI portfolio JSON here]

You are an institutional credit analyst comparing two or more ACI Risk Indicators. For each provider, extract from aci_decision_object:

- Score and band
- Top ceiling factor (criterion_label + point_impact + detail)
- Top floor factor (criterion_label + point_impact + detail)
- Hard caps fired
- Band gap distance

Render a comparison table. Then in 2-3 sentences identify which provider has the strongest risk profile and which has the weakest, citing only aci_decision_object fields. Do not extrapolate beyond the JSON.

[paste 2+ ACI score JSONs here]

You are evaluating an allocation scenario against ACI scores. For the proposed allocation (capital amounts, providers, weights) and the provider score JSONs below, identify:

1. Which providers carry hard caps and what those caps imply for the allocation.
2. The largest ceiling-factor exposure across the portfolio (sum of weighted point_impacts).
3. The points_to_next_band gap for the worst-scored holding.
4. Any concentration above 30% in providers with negative ceiling-factor point_impacts.

Output 4 bullet findings, then 1 sentence overall posture. Do not propose alternative allocations unless explicitly asked.

[paste scenario + ACI score JSONs here]

MCP Server (Agent Integration)

Connect any MCP-compatible AI agent directly to ACI scoring data. Supports Claude and any agent runtime implementing the Model Context Protocol.

# MCP manifest endpoint (public)
GET https://api.aethoncredit.com/api/mcp

# Tool call endpoint (auth-gated)
POST https://api.aethoncredit.com/api/mcp
x-api-key: sk_live_your_key_here

{
  "tool_name": "get_provider_score",
  "input": { "provider_name": "Coinbase Prime" }
}

Available tools: get_provider_score, compare_providers, get_score_history, search_providers, get_methodology. Registration in the public MCP directory is pending counsel review.

Rate Limits

Example Response

{
  "id": "uuid",
  "name": "Provider Name",
  "planner_type": "btc",
  "final_score": 88,
  "risk_band": "LOW",
  "yield_min": 6.5,
  "yield_max": 8.2,
  "scored_at": "2026-04-09T08:00:00Z",
  "limitations_disclosure": "ACI Framework v1.0 · aethoncredit.com · Quantitative risk analytics outputs for independent analysis"
}

Ready to Integrate?

Get your API key in Aethon Terminal to start building.