API Documentation

Programmatic access to Auggie's AI-powered account research.

Quick Start

pip install auggie

from auggie import AuggieClient

client = AuggieClient("aug_your_key")
result = client.research_and_poll("https://example.com")

print(result["scores"]["composite"])       # Overall score
print(result["scores"]["pain"])            # Pain score
print(result["sections"]["talking_points"]) # Ready-to-use talking points

How it works

Create an API key at /api-keys.
Submit a company URL via POST /v1/research. Returns a job_id immediately.
Poll for results via GET /v1/research/{job_id}, or register a webhook.
Generate outreach via POST /v1/research/{doc_id}/sequence.

Base URL

https://auggie.tools/v1

Response format

Every response includes an object field indicating the resource type. Collections use the standard list envelope:

{
  "object": "list",
  "data": [...],
  "has_more": true
}

See Pagination for details on cursor-based navigation.

Python SDK

The official Python SDK wraps the REST API with polling helpers, automatic retries, and typed responses.

from auggie import AuggieClient, AuggieError

client = AuggieClient("aug_your_key")

try:
    result = client.research_and_poll("https://example.com")
except AuggieError as e:
    print(f"Error {e.status_code}: {e.message}")

Async usage: AsyncAuggieClient provides the same API with await.

Understanding scores

Every research result includes four scores from 0-100, plus a composite score that combines them.

Score	Weight	What it measures
pain	40%	How much pain the company is likely experiencing
fit	35%	How well the company matches your ICP
timing	25%	Signals the company is ready to buy now
composite	—	Weighted combination of all scores

Score calibration

Range	Interpretation
85-100	Very high. Strong, clear signals detected. Prioritize these accounts.
70-84	High. Good signals detected. Worth pursuing.
50-69	Moderate. Some signals present. May need qualification.
0-49	Low. Weak or absent signals.