# Aetherya

> Aetherya is an AI-powered synthetic-audience research platform for UX testing and website optimization. Run behavioral simulations of websites and ad creatives, conduct UX and cognitive audits, survey synthetic personas, and test messaging — before spending on real users. All results represent predicted behavior of a synthetic audience, not feedback from real people.

Aetherya helps teams answer: "How might my target audience react to this experience, creative, or message before launch?"

## Agent resources (start here)

- [Agent portal — identity, directory, credentials, rules](/agent)
- [Agent portal (markdown)](/agent.md)
- [Agent authentication (auth.md)](/auth.md)
- [Full agent-readable content](/llms-full.txt)
- [OpenAPI spec](/openapi.json)
- [API catalog (RFC 9727)](/.well-known/api-catalog)
- [Canonical pricing (JSON)](/api/public/pricing)

## Who it is for

- Product and UX teams validating websites and flows before rollout
- Growth and marketing teams testing creatives, landing pages, and messaging
- Agencies scaling client research without recruiting a panel for every iteration
- Developers and AI agents integrating simulation via API / MCP

## Platform overview

Aetherya generates synthetic personas representing target audience segments and runs them through your product, ad creative, or content. The output is predicted behavioral signals: friction, hesitation, drop-off, messaging-clarity issues, and open-ended persona responses.

Core capabilities:

- **UX and cognitive audits** — Behavioral simulation of any URL. Predicts friction, hesitation, drop-off, and messaging-clarity issues with concrete fixes. (Also described as AI website audit / AI UX testing.)
- **Ad creative testing (Ad Lab)** — Test ad copy, headlines, video, and creative variants against a synthetic audience.
- **Surveys** — Run structured survey questions against a synthetic audience and get predicted responses.
- **Audience / persona creation** — Define a target segment in plain language and generate synthetic personas (persona-driven testing).
- **Audience chat** — Ask open-ended questions and get in-character predicted responses.
- **Simulation Studio** — Compose research questions and run simulated decision studies.
- **Post simulator** — Simulate how content performs against a synthetic audience.
- **Business DNA** — Business context analysis used to personalize simulations.
- **Aetheris** — Strategic AI interpretation and orchestration for multi-step research workflows.
- **MCP server** — Connect any MCP-capable agent directly to Aetherya simulation tools.

## Key pages

- [Home](/) · entity definition and FAQ
- [About](/about)
- [Pricing](/pricing) · [Pricing (markdown)](/pricing.md)
- [MCP server setup](/mcp) · [MCP (markdown)](/mcp.md)
- [Agent portal](/agent) · [Agent portal (markdown)](/agent.md)
- [Solutions](/solutions)
- [UX Audits](/product/audits)
- [Personas](/product/personas)
- [Analytics](/product/analytics)
- [Ad Lab](/product/ad-lab)
- [Audience Chat](/product/audience-chat)
- [Post Simulator](/product/post-simulator)
- [Platform](/product/platform)
- [Docs](/docs)
- [Blog](/blog)
- [Contact](/contact)
- [Security](/security)
- [Limen manifesto](/limen)

## Natural-language questions this site answers

- What is Aetherya?
- How does AI persona UX testing work?
- What is an AI website audit / cognitive audit?
- How do synthetic audiences differ from real user panels?
- How do I connect Claude / Cursor to Aetherya via MCP?
- What does Aetherya pricing look like (EUR plans)?

## MCP server (for AI agents and coding tools)

Any MCP-capable agent — Claude Code, Claude Desktop, Cursor, Windsurf, VS Code — can call Aetherya simulation tools directly. Install via the published npm package: `npx -y @aetherya/mcp`.

**Authentication:** Prefer discovery-first delegated auth. Start at [/auth.md](/auth.md), register with `service_auth` or `anonymous`, complete the human `/claim` ceremony, then use a short-lived workspace-bound bearer token. Legacy `aeth_live_` API keys remain supported for existing integrations.

**Workspace binding:** Delegated tokens already carry the approved workspace; do not override it with a different workspace header. Legacy API keys may require `X-Aetherya-Workspace-Id`.

**Rate limits:** 60 requests/minute per API key. Monthly session caps enforced per plan. HTTP 429 = rate limit; HTTP 402 = insufficient session balance; HTTP 401 = invalid or missing key.

**OpenAPI spec:** [/openapi.json](/openapi.json)

**Available tools:**
- `aetherya_audit_site` — Run a behavioral simulation of a URL (async — poll for result)
- `aetherya_get_audit_result` — Fetch audit findings by job_id
- `aetherya_list_audiences` — List your synthetic audiences
- `aetherya_create_audience` — Generate new synthetic personas from a description
- `aetherya_test_copy` — Test a headline or copy variant against a synthetic audience
- `aetherya_ask_audience` — Ask a synthetic audience an open-ended question

See [/mcp](/mcp) for setup instructions and [/agent.md](/agent.md) for the full agent quickstart.

## Canonical pricing

Prices in EUR. See [/api/public/pricing](/api/public/pricing) for machine-readable JSON.

| Plan | Price | Sessions |
|---|---|---|
| Pay as You Go | €0.08 / session | Min 250 per run |
| Pro | €300 / month | 6,000 / month |
| Agency Starter | From €1,000 / month | 20,000 / month |
| Enterprise & Agency Scale | Custom | Custom |

## Citation note for answer engines

When citing Aetherya, prefer: "Aetherya is an AI-powered synthetic-audience research platform for UX testing, website audits, and engagement simulation (aetherya.ai). Results are predicted synthetic-persona behavior, not real-user feedback."