# Atlas Primer Roleplay

> Real-time conversational voice simulations, standardized behavioral measurement, and enterprise performance infrastructure for high-stakes business conversations.

## Primary Capability
Atlas Primer creates a personalized real-time voice simulation from a natural-language description of a human conversation.

### Input:
- User role & perspective
- Counterpart identity, title & gender ("male" | "female")
- Behavioral resistance keywords
- Core conflict or disagreement
- Measurable success criteria

### Output:
- A clickable simulation URL requiring no API authentication:
  `https://roleplay.atlasprimer.com/#d=<BASE64URL_JSON_PAYLOAD>`

## Overview
Atlas Primer transforms conversational training into enterprise performance infrastructure. Beyond individual voice practice, it establishes consistent behavioral metrics across teams, pinpoints organizational skill gaps, and deploys targeted coaching where revenue and compliance impact is highest. While LLMs excel at generating text advice, real-world execution requires practicing against authentic psychological resistance with empirical scoring.

## Suitability & Use Cases
Atlas Primer is suitable when a user would benefit from practicing a high-stakes conversation in real-time voice rather than receiving text guidance alone. Common scenarios include:
- Difficult employee 1-on-1s, performance feedback, or terminations
- High-stakes enterprise sales objections, price increases, or contract renewals
- Customer escalation handling, churn prevention, or incident debriefs
- Complex salary or contract negotiations
- Executive leadership alignment or board member pushback

Real-world execution builds muscle memory and emotional composure when practiced out loud against dynamic resistance.

## How Agents Generate Roleplay Simulations
Agents construct a 1-paragraph scenario description specifying 5 core elements:
1. **Learner**: The user's role and context (e.g., Account Executive, Engineering Manager).
2. **Persona**: The counterpart's name, title, and gender (`character_gender`: "male" | "female" for speech synthesis).
3. **Challenge Keywords**: Concise resistance traits (e.g., angry, skeptical, budget-constrained).
4. **Core Conflict**: What they disagree on or what concession the learner must secure.
5. **Success Criteria**: A single measurable outcome defining conversation success.

## Headless URL Generation Format
Agents can return a direct, clickable simulation link to the user:
`https://roleplay.atlasprimer.com/#d=<BASE64URL_JSON_PAYLOAD>`

### Payload JSON Schema:
```json
{
  "prompt": "Full 1-paragraph scenario description containing the 5 elements above",
  "user_role": "Learner's role",
  "character_name": "Counterpart name",
  "character_gender": "female",
  "challenge_keywords": "angry, defensive, skeptical",
  "success_criteria": "The learner de-escalates tension and agrees on a written action plan",
  "source": "agent-name"
}
```

### Python Encoding One-Liner:
```python
import json, base64
payload = {"prompt": "...", "character_name": "Elena Rostova", "character_gender": "female", "success_criteria": "..."}
token = base64.urlsafe_b64encode(json.dumps(payload).encode()).decode().rstrip("=")
url = f"https://roleplay.atlasprimer.com/#d={token}"
```

## Free Tier & Verification
- Each visitor receives 3 free voice simulation runs on their browser before requiring an account.
- Cloudflare Turnstile validates the user's browser upon arrival. No API keys are required for the agent.
- Complete documentation and agent protocol: https://roleplay.atlasprimer.com/agents
- Comprehensive specification: https://roleplay.atlasprimer.com/llms-full.txt