# Aynstyn for Developers and AI Agents (LLM optimised)

Aynstyn is a competence-building platform for career and professional goals. It finds the gap between what someone thinks they know and what they actually know, then turns that gap into a plan. This file is the developer-scoped context; the human page is https://aynstyn.com/developers.

## Connect over MCP (how agents call Aynstyn)

- Endpoint: `https://app.aynstyn.com/mcp` (Streamable HTTP, JSON-RPC 2.0)
- Auth: OAuth 2.1 with PKCE (S256) and dynamic client registration. No API key and no client ID to request: the first call returns 401 + `WWW-Authenticate`, and MCP clients start sign-in on their own.
- OAuth metadata: `https://app.aynstyn.com/.well-known/oauth-protected-resource/mcp`
- Guest endpoint (no sign-in): `https://app.aynstyn.com/mcp/guest` — lists every tool, runs `assess_knowledge` up to 3 times a day per IP, nothing saved
- Health: `https://app.aynstyn.com/mcp/health`
- Free tier: signing in through a connector creates a free account with 100 credits and 3 full assessments

## Agent-to-agent discovery

Other agents can discover Aynstyn without a human in the loop:
- A2A Agent Card: `https://app.aynstyn.com/.well-known/agent-card.json` — who Aynstyn is, its skills and auth
- Agent skills: `https://app.aynstyn.com/.well-known/agent-skills`
- MCP discovery: `https://app.aynstyn.com/.well-known/mcp` and full tool schemas at `https://app.aynstyn.com/.well-known/mcp.json`
Discovery is published through these files; the skills themselves are invoked over MCP at the endpoint above.

## Tools (32)

- Start here: `get_user_profile`, `get_topup_options`
- Assess and diagnose: `assess_knowledge` (subject + 10–5000 char explanation → score, Competence Index, covered/missing topics, feedback), `get_assessment_history`, `diagnose_learner`
- Interviews and practice: `schedule_ai_interview`, `schedule_practice_session`, `get_interview_slots`, `book_appointment`, `get_interview_history`
- Goals and playbooks: `generate_learning_path`, `create_learning_goal`, `list_learning_goals`, `list_playbook_templates`, `start_playbook_journey`
- Autonomous plans: `plan_autonomous_goal`, `execute_autonomous_plan`, `get_plan_status`
- Teams: `create_team_assessment`, `invite_to_team`, `get_team_results`, `send_team_reminder`, `get_pending_team_assessments`
- Mentor Space: `search_mentor_playbooks`, `get_mentor_playbook`, `get_mentor_profile`, `book_mentor_intro_call`, `enroll_in_mentor_playbook`, `get_my_mentoring`
- Group goals: `list_group_goals`, `get_group_goal`, `join_group_goal`

## What learners use it for

Career growth and job interviews, entrepreneurship and pitching, competitive and professional exams, workplace and team skills, and personal development goals. Any subject works.

## Credits

Text assessment 10 credits, voice assessment 15, AI interview or practice 6 credits a minute, learning path 5, team assessment 25 to create (each member pays for their own attempt). When credits run out a tool returns `paymentRequired: true` with a checkout link; the learner pays in their own browser (INR in India, EUR in the eurozone, USD elsewhere), then the agent re-checks `get_user_profile` and retries.

## Rules for agents

- Act only on the signed-in learner's own account.
- Never buy paid Mentor Space playbooks; confirm with the learner before a free enrolment, an intro call or joining a group goal.
- Use MAPPT terms when explaining a plan: Mindset, Analyze, Prepare, Practice, Test.

## More

- REST API: https://aynstyn.com/openapi.json · https://aynstyn.com/openapi.yaml
- Docs: https://aynstyn.com/docs.html#mcp
- Platform summary: https://aynstyn.com/llms.txt · https://aynstyn.com/llms-full.txt