# Aetherya (npm · @aetherya/mcp)

Test decisions on a synthetic audience: studies, site audits, copy tests and post simulations.

- Trust score: 68/100 (medium)
- Change this week: +4
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `mcp.aetherya.ai`: 72/100, [markdown](https://verifymcp.io/servers/ai-aetherya-mcp/mcp.md), [page](https://verifymcp.io/servers/ai-aetherya-mcp/mcp)
- npm · `@aetherya/mcp`: 68/100 (this document), [markdown](https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp.md), [page](https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp)

## Channel facts

- Registry: `npm`
- Package: `@aetherya/mcp`
- Version: `1.0.5`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Supply Chain Security**: 81/100
  - No malware found by supply-chain analysis.
  - CVE check failed: a known high-severity CVE affects zod 3.25.76, a direct dependency. A fixed version is available.
  - No install/post-install scripts declared.
  - 30 of 93 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 19/100
  - Repository check failed: no source repository is declared.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 16 days ago).
  - Security-disclosure policy not yet verified: we couldn't inspect the source repository.
- **Schema Quality & AI Usability**: 75/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 7715 tokens (~285/item across 27 items; 27 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 57/100
  - Stability observed for 17 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 27 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 28 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the Aetherya MCP server?

Aetherya runs locally as an npm package, launched with npx -y @aetherya/mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add ai-aetherya-mcp -- npx -y @aetherya/mcp
```

### Cursor

```json
{
  "mcpServers": {
    "ai-aetherya-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@aetherya/mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-aetherya-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@aetherya/mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add ai-aetherya-mcp -- npx -y @aetherya/mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-aetherya-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@aetherya/mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-aetherya-mcp --command npx --arg -y --arg @aetherya/mcp
```

### Hermes

```yaml
mcp_servers:
  ai-aetherya-mcp:
    command: "npx"
    args: ["-y", "@aetherya/mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "ai-aetherya-mcp": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "@aetherya/mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-aetherya-mcp -t stdio -c npx -a -y @aetherya/mcp
```

### Other

```json
{
  "mcpServers": {
    "ai-aetherya-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@aetherya/mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-05 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 53 to 57. That category is still filling its 30-day observation window: 16 days of observed history at the previous scan, 17 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-03 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 47 to 50. That category is still filling its 30-day observation window: 14 days of observed history at the previous scan, 15 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-01 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-30 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 64, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 63, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 62, −5)

- [security regression] CVE-2023-54404 affects this package: high
- [security regression] Known CVEs: pass → fail

### 2026-10-23 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 10 to 13. That category is still filling its 30-day observation window: 3 days of observed history at the previous scan, 4 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (27)

### `aetherya_list_capabilities` (~126 tokens)

List what you can run and what plan headroom is left

List the research methods this workspace can run — with the exact stimulus fields each one requires — plus remaining plan headroom and simulation-session balance. Call this before aetherya_run_study or aetherya_test_copy to pick a method and check cost headroom. Everything Aetherya runs is a SIMULATION of a synthetic audience, not real-user research.

Input parameters:

- `section` (string): 'methods' lists the research methods available and the inputs each needs. 'limits' returns just your remaining plan headroom and session balance. Default 'all'.

Output parameters:

- `methods` (array)
- `notes` (array)
- `plan`
- `source` (string)
- `tools` (array)

### `aetherya_estimate` (~159 tokens)

Estimate the cost of a simulation before running it

Quote how many simulation sessions a run would cost, and roughly what share of your plan it would consume, before committing to it. Free — this never starts a simulation. Use it to size an audience or shorten a study when headroom is tight.

Input parameters:

- `audience_size` (integer): Simulated respondents / audience size.
- `depth` (string): Engagement depth. Cost multiplier: 1x / 2x / 4x.
- `kind` (string, required): What you are about to run.
- `method` (string): Required when kind is 'study' — the research method id.
- `questions` (integer): Number of survey questions.
- `variants` (integer): Number of copy variants, for a copy test.

Output parameters:

- `breakdown`
- `eurEquivalent` (number)
- `kind` (string)
- `note` (string)
- `planImpact`
- `sessions` (number|null): Estimated simulation sessions, or null when Aetherya cannot quote this kind up front. Never treat null as zero — read `note` instead.
- `source` (string)

### `aetherya_plan_study` (~413 tokens)

Plan and run a study from a plain-English objective

Give Aetherya a decision you need to make, in plain English, and it designs the study, picks or builds the audience, prices it, and runs it — then writes a decision brief. Use this when you do not already know which research method fits; use aetherya_run_study when you do. It can also simulate a whole funnel (an ad leading to a landing page, and which call to action people take there): describe the ad, the page URL and the goals in the objective. It always stops for your approval at a price before spending anything: expect status 'needs_price_confirmation' and answer it with aetherya_confirm_study. Call it again with autopilot_id to check on a plan already running. Results are SIMULATED predicted behaviour, not real-user research.

Input parameters:

- `asset_ids` (array): Optional. Ids of ads the user already uploaded to Aetherya (images or videos). Pass them when the objective is about those ads, e.g. a funnel: 'our Instagram ad leads to example.com/offer; do people…
- `audience_id` (string): Optional saved audience pack / persona group id (UUID) to draw the sample from.
- `autopilot_id` (string): Check an existing plan instead of starting one. Mutually exclusive with objective.
- `language` (string): Optional. Report language, e.g. 'ro'.
- `market` (string): Optional. Target market, e.g. 'RO'.
- `max_wait_seconds` (integer): How long to wait inline before handing back an id. Defaults to 45.
- `objective` (string): What you need to decide, in plain English — e.g. 'should we price the starter plan at 29 or 49 for early-stage dev teams?'. Starts a new plan. Omit when checking an existing one.
- `register` (string): Optional. Report tone register.

Output parameters:

- `audience` (string)
- `autopilotId` (string)
- `brief`
- `error` (string)
- `estimate`
- `id` (string)
- `kind` (string)
- `legs` (array)
- `next` (string)
- `progress` (array)
- `question` (string)
- `questions` (array)
- `research` (string)
- `source` (string)
- `status` (string): 'running' | 'needs_price_confirmation' | 'needs_clarification' | 'completed' | 'failed' | 'cancelled'. The two 'needs_' values are states with a concrete next action, not failures.
- `step` (string)
- `stepLabel` (string)

### `aetherya_confirm_study` (~336 tokens)

Approve, answer, cancel or retry a planned study

Respond to a plan from aetherya_plan_study. Use action 'confirm' with accept_sessions to approve the quoted cost and start the run — this is the point at which simulation sessions are spent, so check with the user first. Use 'answer' to reply to the clarifying questions (answers with the option ids the user picked, or answer for one free-text reply), 'cancel' to abandon the plan, or 'retry' after a failure. After confirming, the run continues and returns its decision brief when it finishes.

Input parameters:

- `accept_sessions` (number): Required for 'confirm': the exact sessions figure from the plan's estimate. Never invent it — a mismatch is rejected and re-quoted.
- `action` (string, required): 'confirm' accepts the quoted cost and starts the run (requires accept_sessions). 'answer' replies to the clarifying questions (requires answers or answer). 'cancel' abandons the plan. 'retry' re-runs…
- `answer` (string): For 'answer': a free-text reply to the (first) question asked. Prefer answers when the plan lists questions.
- `answers` (array): For 'answer': one entry per question from the plan's questions, with the optionId the user picked or their own text. A question left out keeps its default. Only relay what the user said.
- `autopilot_id` (string, required): The autopilotId from aetherya_plan_study.
- `max_wait_seconds` (integer): How long to wait inline after confirming before handing back an id. Defaults to 45.

Output parameters:

- `audience` (string)
- `autopilotId` (string)
- `brief`
- `error` (string)
- `estimate`
- `id` (string)
- `kind` (string)
- `legs` (array)
- `next` (string)
- `progress` (array)
- `question` (string)
- `questions` (array)
- `requiredSessions` (number): Set only on a re-quote — confirm this figure instead.
- `research` (string)
- `source` (string)
- `status` (string)
- `step` (string)
- `stepLabel` (string)

### `aetherya_read` (~345 tokens)

Instant read: which variant lands with a market

Myrias instant read. Reads 2-5 variants against a census-weighted synthetic population for a market, in seconds, and returns each variant's predicted action propensity, its lift over the control, and which demographic archetypes favour it. A variant can be copy (text), a live page (url), an image (image_path, image_url or image_base64) or a video (video_path). Media reads are single-kind: images or videos cannot be mixed with copy, pages or each other. The first read of a new image takes about 30-45 s and may come back running (collect it with aetherya_get_result). A video is uploaded and prepared first, about 1-2 min. Much cheaper and faster than aetherya_test_copy, which is the deeper follow-up. Counts against the monthly read allowance, not simulation sessions. Results are SIMULATED, not real-user data.

Input parameters:

- `audience` (string): Read against a saved audience (persona group id from aetherya_list_audiences) instead of a market.
- `market` (string): Who to read against, as a place: 'United States', 'Germany', 'RO'. Defaults to the US.
- `target_action` (string): The action the variants should drive, e.g. 'Sign up'.
- `title` (string): Optional name for the read in your run history.
- `variants` (array, required): 2-5 variants to read side by side. Images and videos are read on their own, not mixed with other kinds.
- `where` (string): Where the variants appear. Defaults to 'Landing page'.

Output parameters:

- `archetypes` (array)
- `explanation` (string)
- `next` (string)
- `population`
- `runId` (string)
- `source` (string)
- `status` (string): 'completed' | 'running' | 'failed'.
- `studyId` (string)
- `validation` (string)
- `variants` (array)

### `aetherya_run_study` (~473 tokens)

Run a research study with a synthetic audience

Run a Research Lab study — pricing sensitivity, message testing, concept testing, focus groups — against a synthetic audience and get a structured verdict with segment breakdowns, objections and representative reactions. Call aetherya_list_capabilities first to pick a method and see the inputs it needs. Results are SIMULATED predicted behaviour, not feedback from real users.

Input parameters:

- `accept_sessions` (number): Only needed after a 'needs_price_confirmation' result. Pass back the exact requiredSessions value from that response to proceed.
- `audience` (string): Optional saved audience (persona group) id to draw the sample from. Omit to draw from your existing synthetic audience.
- `audience_size` (integer): Simulated respondents. Defaults to the method's own default (usually 30). Drives cost.
- `depth` (string): How deeply each respondent engages. Cost multiplier: single_pass 1x, deliberation 2x, deep 4x.
- `hypothesis` (string): Optional. What you expect to find.
- `language` (string): Optional. Report language, e.g. 'ro'.
- `market` (string): Optional. Target market, e.g. 'RO'.
- `max_wait_seconds` (integer): How long to wait inline before handing back an id to poll. Defaults to 45.
- `method` (string, required): Research method id, from aetherya_list_capabilities — e.g. 'messaging_test', 'concept_test', 'feature_adoption', 'pricing_van_westendorp', 'focus_group'.
- `objective` (string, required): The decision this study should inform.
- `register` (string): Optional. Report tone register.
- `regulated` (boolean): Optional. True for regulated sectors (claims are checked).
- `scenario` (string): Optional emotional condition from the method's supportedConditions, e.g. 'budget_anxiety'.
- `stimulus` (object, required): The method's own stimulus fields, keyed by the `key` values in that method's stimulusSchema (see aetherya_list_capabilities). A variant_list field takes an array of { text, isControl } objects, 2-5 e…
- `title` (string, required): Short name for this study, e.g. 'Q3 pricing band for Pro'.

Output parameters:

- `adoption`: synthesis.adoption (feature_adoption method).
- `claimFlags` (array): synthesis.claimFlags: regulated-claim warnings.
- `confidence`
- `emotionalDistribution` (array)
- `eurEquivalent` (number)
- `methodId` (string)
- `next` (string): Suggested next call.
- `objections` (array)
- `priceCurve`
- `recommendedNext` (array)
- `requiredSessions` (number): Present on 'needs_price_confirmation'.
- `runId` (string)
- `sampleSummary`
- `segmentBreakdown` (array)
- `sessionsConsumed` (number): Simulation sessions billed to this run.
- `sessionsEstimated` (number)
- `source` (string)
- `status` (string): 'completed' | 'running' | 'failed' | 'needs_price_confirmation' | 'needs_audience'.
- `studyId` (string)
- `variantComparison`
- `verbatims` (array)
- `verdict`

### `aetherya_test_copy` (~238 tokens)

Test copy variants against a synthetic audience

Compare 2-5 copy variants — headlines, value propositions, ad or email copy — against a synthetic audience, and get a ranked verdict with predicted action propensity, comprehension and trust per variant, plus the objections each raises. Results are SIMULATED predicted reactions, not feedback from real users.

Input parameters:

- `accept_sessions` (number): Only after a 'needs_price_confirmation' result: the exact requiredSessions value from that response, re-sent to authorize the spend. Never guess it.
- `audience` (string): Optional saved audience (persona group) id to test against.
- `audience_size` (integer): Simulated respondents. Default 30. Drives cost.
- `max_wait_seconds` (integer): How long to wait inline before handing back an id to poll. Defaults to 45.
- `objective` (string): What this copy should achieve, e.g. 'get developers to start a free trial'.
- `target_action` (string): The action the copy should drive.
- `variants` (array, required): 2-5 copy variants to compare head-to-head.
- `where` (string): Where this copy appears.

Output parameters:

- `eurEquivalent` (number)
- `next` (string)
- `objections` (array)
- `requiredSessions` (number)
- `runId` (string)
- `sessionsConsumed` (number)
- `sessionsEstimated` (number)
- `source` (string)
- `status` (string): 'completed' | 'running' | 'failed' | 'needs_price_confirmation' | 'needs_audience'.
- `studyId` (string)
- `variantComparison`
- `verbatims` (array)
- `verdict`

### `aetherya_audit_site` (~678 tokens)

Audit a site with a synthetic audience

Run a behavioral simulation: a synthetic audience visits a URL and predicts where real users would hesitate, get confused, or drop off, plus messaging-clarity and UX friction findings with concrete fixes. Ideal for evaluating a freshly built or changed page and getting actionable UX/messaging improvements in the same loop. To find out how people experience a flow (sign-up, onboarding, checkout), pass tasks and start_url: the simulated visitors then walk the flow step by step instead of touring the site. By default they stop before the final submit, so nothing is created or bought; pass allow_test_signups only when the user explicitly allows test sign-ups. Payment is never submitted. For pages behind a login, pass credentials for a dedicated test account the user gave you. Results are SIMULATED (predicted behavior of a synthetic audience), not feedback from real users.

Input parameters:

- `allow_test_signups` (boolean): Flow tests only. Default false: visitors stop before the final submit, so no account is created and no order is placed. Set true only if the user explicitly allows test sign-ups on this site. Payment…
- `audience` (string): Optional. A synthetic audience to use: a persona id or name from aetherya_list_audiences, or a short free-text description (e.g. 'first-time SaaS buyers'). If omitted, Aetherya simulates a small mixe…
- `credentials` (object): Use a dedicated test account. Sent once, stored encrypted, deleted when the run ends. Never use a real customer or admin account.
- `focus` (string): Optional. What to pay attention to, e.g. 'checkout flow', 'onboarding clarity', 'is the value proposition clear?'.
- `max_wait_seconds` (integer): How long to wait inline before handing back an id to poll. Defaults to 45.
- `num_agents` (integer): How many simulated visitors walk the page. Default 3. Drives cost — quote it with aetherya_estimate (kind 'audit', audience_size = this number) before raising it.
- `objective` (string): Optional. What you want to find out, in your own words, e.g. 'how do new users feel about our onboarding flow?'. Sent as written (not paraphrased); steers the simulated visitors and the report.
- `scope` (object): Optional. Restrict where visitors may go: 'site' (whole domain, default), 'path' (a subtree, value '/blog') or 'page' (one page, value a path or full URL).
- `start_url` (string): Optional. Where a flow test begins, e.g. https://example.com/signup. Same site as url. Use together with tasks.
- `tasks` (array): Optional flow test (max 5 tasks). Each simulated visitor walks the flow step by step and reports where they stall and how each step felt, instead of touring the site. Use it for 'how do people experi…
- `url` (string, required): The full URL of the page to audit (e.g. https://example.com/pricing).

Output parameters:

- `executiveSummary` (string)
- `findings` (array)
- `frictionPoints` (array)
- `jobId` (string)
- `next` (string): Suggested next action when status is 'running'.
- `personaBehavior` (array)
- `personaJourneys` (array)
- `recommendations` (array)
- `reportId` (string)
- `reportLanguage` (string)
- `score` (number): Predicted experience score, 0-100, when available.
- `source` (string): Always 'synthetic_audience_simulation'.
- `status` (string): 'completed' when findings are inline, or 'running' when pollable.
- `url` (string)

### `aetherya_test_flow` (~367 tokens)

Test which exit an ad and landing page drive people to

Simulate people who arrive from an ad (image, video or copy), land on a page, and take one of several exits (for example book a consultation, open the plan form, or leave). Reports the exit split per arm. Pass rewrites to compare a changed ad and page copy against the original on the same audience. The first call quotes the cost and starts nothing; call again with accept_cost true once the user agrees. Collect results with aetherya_get_result kind 'flow'. Results are SIMULATED, not real-user data.

Input parameters:

- `accept_cost` (boolean): Leave unset on the first call: nothing starts and the quote comes back. Set true only after the user agrees to the quoted cost.
- `ads` (array): 0-2 ads. Each simulated visitor sees one ad first; with two, they are split across them.
- `audience` (string): Optional. A synthetic audience: a persona id or name from aetherya_list_audiences, or a short free-text description (e.g. 'first-time SaaS buyers').
- `exits` (array, required): 1-4 ways the visit can end. A visitor takes the first exit whose signal matches, or leaves. Give at least one signal per exit: form_contains (most reliable), url_contains or text_visible.
- `num_agents` (integer): Simulated visitors per arm. Default 6. Each arm runs this many, and the ads are split round-robin across them. Cost is quoted per arm.
- `page_url` (string, required): The landing page the ads send people to.
- `rewrites` (object): Adds a second arm, 'rewrite', run against the same audience as 'original'. Give ads and/or page_copy.

Output parameters:

- `armCount` (number)
- `arms` (array)
- `estimatedPercentMax` (number): Plan share for all arms, high end.
- `estimatedPercentMin` (number): Plan share for all arms, low end.
- `flowRunId` (string)
- `next` (string)
- `numAgents` (number)
- `source` (string)
- `status` (string): 'needs_price_confirmation', 'running', or 'error'.

### `aetherya_ask_audience` (~292 tokens)

Ask ONE synthetic persona a question (secondary)

SECONDARY TOOL. Pose a question to a SINGLE synthetic persona and get one predicted, in-character response — good for an open-ended 'would this confuse someone like this?' sounding-out. The answer is SIMULATED (one persona's predicted reaction), not a real person, and ONE persona is not an audience: it carries no distribution and no disagreement, so never report it as what 'users' or 'the audience' think. For a finding you can act on, use aetherya_test_copy (compare variants) or aetherya_run_study / aetherya_plan_study (a panel with a spread of reactions).

Input parameters:

- `audience` (string): Optional. A persona id or name from aetherya_list_audiences to answer as. If omitted, one of your synthetic audiences is chosen automatically.
- `context` (string): Optional. Background/situation for the audience to consider when answering.
- `language` (string): Optional. The language to answer in, e.g. 'English' or 'Romanian'. Personas answer in their own native language by default, which is correct for research but unreadable when you asked in another lang…
- `question` (string, required): The question to ask the synthetic audience (e.g. 'Would this pricing page confuse you?').

Output parameters:

- `answer` (string)
- `audience` (object)
- `personaCount` (number)
- `question` (string)
- `sessionId` (string)
- `source` (string)

### `aetherya_simulate_post` (~267 tokens)

Simulate how an audience reacts to a post

Run a post (X, LinkedIn, Instagram, Facebook) or email through a synthetic audience and get predicted engagement, sentiment and representative reactions before publishing. Starts the run and waits briefly; collect with aetherya_get_result kind 'post_simulation'. Results are SIMULATED reactions, not real engagement.

Input parameters:

- `accept_sessions` (number): Only after a 'needs_price_confirmation' result: the exact requiredSessions value from it, re-sent to authorize the spend. Never guess it.
- `audience_size` (integer): Simulated audience members. Default 100. Drives cost.
- `image_url` (string): Public URL of an image attached to the post, if any.
- `link_url` (string): Link attached to the post, if any.
- `objective` (string): What the post is for, e.g. 'drive webinar signups'.
- `persona_ids` (array): Specific personas (from aetherya_list_audiences) to use. Omit to use the workspace pool.
- `platform` (string, required): Where the post goes out.
- `text` (string, required): The post (or email body) exactly as it would be published.
- `tone` (string): Tone of the simulated comments. Default balanced.

Output parameters:

- `audienceSize` (number)
- `executiveSummary` (string)
- `id` (string)
- `kind` (string)
- `next` (string)
- `platform` (string)
- `reactions` (array)
- `requiredSessions` (number)
- `source` (string)
- `status` (string): 'running' | 'completed' | 'needs_price_confirmation' | 'needs_audience'.

### `aetherya_cancel` (~102 tokens)

Cancel a running Aetherya run

Stop a study, instant read, plan, audit or post simulation that is still running. Use it when the user changes their mind or a run was started with the wrong inputs. Finished runs are not affected.

Input parameters:

- `id` (string, required): run_id for a study or read, autopilotId for a plan, job_id for an audit, or the post simulation id.
- `kind` (string, required): The kind named in the run tool's response.

Output parameters:

- `id` (string)
- `kind` (string)
- `source` (string)
- `status` (string)

### `aetherya_list_runs` (~108 tokens)

List recent Aetherya runs

List the workspace's recent simulations across studies, instant reads, plans, audits and post simulations, newest first, with the kind and id to pass to aetherya_get_result or aetherya_get_brief. Use it to find earlier work instead of re-running it.

Input parameters:

- `kind` (string): Only this kind of run. Default 'all'. Studies include copy tests and instant reads.
- `limit` (integer): How many to return. Default 15.

Output parameters:

- `runs` (array)
- `source` (string)
- `unavailable` (array)

### `aetherya_get_result` (~226 tokens)

Get the result of an Aetherya run

Fetch the findings of a run started by aetherya_run_study, aetherya_test_copy, aetherya_audit_site, aetherya_plan_study, aetherya_read, aetherya_simulate_post, aetherya_compare_creatives, aetherya_pretest_video or aetherya_test_flow, using the kind and id from that tool's response. Simulations run longer than a single tool call, so this is the normal way to collect results. All results are SIMULATED predicted behaviour, not real-user data.

Input parameters:

- `id` (string, required): The id from the run tool — job_id for audits, run_id for studies and copy tests, autopilotId for plans, the simulation id for a post simulation, the comparisonId for a creative comparison, the experi…
- `kind` (string, required): Which kind of run to fetch. Use the kind named in the run tool's `next` hint.

Output parameters:

- `adoption`: synthesis.adoption (feature_adoption method).
- `archetypes` (array)
- `arms` (array)
- `attentionCurve` (array)
- `audience` (string)
- `audienceSize` (number)
- `autopilotId` (string)
- `brief`
- `claimFlags` (array): synthesis.claimFlags: regulated-claim warnings.
- `concepts`
- `confidence`
- `dropOffHotspots` (array)
- `emotionalDistribution` (array)
- `error` (string)
- `estimate`
- `executiveSummary` (string)
- `explanation` (string)
- `findings` (array)
- `glance`
- `hookScore` (number|null)
- `hookWindowSec` (number|null)
- `id` (string)
- `kind` (string)
- `legs` (array)
- `methodId` (string)
- `next` (string)
- `objections` (array)
- `personaJourneys` (array)
- `platform` (string)
- `population`
- `priceCurve`
- `progress` (array)
- `question` (string)
- `ranking` (array)
- `reactions`
- `recommendations` (array)
- `recommendedNext` (array)
- `reportId` (string)
- `research` (string)
- `sampleSummary`
- `score` (number)
- `segmentBreakdown` (array)
- `segmentBreakdownBySegment` (array)
- `sentimentBreakdown`
- `sessionsConsumed` (number)
- `source` (string)
- `status` (string): 'completed', 'running', or 'failed'.
- `step` (string)
- `stepLabel` (string)
- `topStrengths` (array)
- `topWeaknesses` (array)
- `url` (string)
- `validation` (string)
- `variantComparison`
- `variants` (array)
- `verbatims` (array)
- `verdict`
- `winner`
- `winnerBySegment`

### `aetherya_get_audit_result` (~151 tokens)

Get the result of a synthetic-audience audit

Fetch the result of an audit started by aetherya_audit_site, using its job_id. Returns the full simulated behavioral findings once complete, or a 'running' status to poll again. Results are SIMULATED predicted behavior, not real users.

Input parameters:

- `include` (array): Extra detail to fetch alongside the report. 'journeys' is the step-by-step path each simulated visitor took; 'heatmap' is where attention landed per page. Both are large — ask for them only when the…
- `job_id` (string, required): The job_id returned by aetherya_audit_site (or an audit report id).

Output parameters:

- `executiveSummary` (string)
- `findings` (array)
- `frictionPoints` (array)
- `heatmap`
- `jobId` (string)
- `journeys`
- `next` (string)
- `personaBehavior` (array)
- `personaJourneys` (array)
- `recommendations` (array)
- `reportId` (string)
- `reportLanguage` (string)
- `score` (number)
- `source` (string)
- `status` (string): 'completed', 'running', or 'failed'.
- `url` (string)

### `aetherya_get_brief` (~285 tokens)

Get the plain-language decision brief for a finished run

Turn a finished study, audit, read or post simulation into a short decision brief: the question, a direct answer, the reasoning with supporting simulated reactions, the risks, and what to do next. This is what you hand a human — use it instead of relaying raw metrics. Reuses the stored brief when one exists, so repeat calls are free (unless language, register or regulated is given: that writes a new brief with a model call). Everything in it comes from SIMULATED behaviour, not real-user research.

Input parameters:

- `id` (string, required): The run id for a study, or the reportId (not the job_id) for an audit.
- `kind` (string, required): 'study' for a run_study / test_copy / plan_study / read run id, 'audit' for an audit report id, 'post_simulation' for a simulate_post id.
- `language` (string): Optional. Brief language (e.g. 'ro'); sent as reportLanguage.
- `regenerate` (boolean): Re-write a brief that already exists. Costs another model call — leave unset to reuse the stored one. Passing language, register or regulated also writes a new brief.
- `register` (string): Optional. Tone register of the brief.
- `regulated` (boolean): Optional. Set true for regulated sectors (claims are checked).

Output parameters:

- `answer` (string)
- `caveat` (string)
- `confidence`
- `doNext` (array)
- `id` (string)
- `kind` (string)
- `next` (string)
- `question` (string)
- `reasons` (array)
- `risks` (array)
- `source` (string)
- `status` (string): 'ready' | 'run_not_complete' | 'generating'.
- `warnings` (array)
- `whatWeRan` (array)

### `aetherya_share_result` (~244 tokens)

Create a public link to a finished run's findings

Mint a shareable URL for a completed study, audit or creative comparison so someone without an Aetherya account can read the findings. Use it after aetherya_get_brief when the user wants to send the results to a colleague or client, or post them. The link is PUBLIC to anyone who has it. A study shared without options gets its public link plus a preview card for posting; pass expires_in or a password for a private, expiring link instead.

Input parameters:

- `expires_in` (string): How long the link stays live. Audits and comparisons default to 30d. A study with no expiry and no password gets its public link (stays live until the owner stops sharing it).
- `id` (string, required): The run id for a study, the reportId (not the job_id) for an audit, or the comparison id.
- `kind` (string, required): 'study' for a research run id, 'audit' for an audit report id, 'creative_comparison' for a compare_creatives comparison id.
- `password` (string): Optional password required to open the link. Never returned in the response.

Output parameters:

- `cardUrl` (string)
- `expiresAt` (string)
- `id` (string)
- `kind` (string)
- `note` (string)
- `passwordProtected` (boolean)
- `source` (string)
- `url` (string)

### `aetherya_pretest_report` (~241 tokens)

Build a pretest report

Assemble a markdown pretest report (how we worked, one section per set, what the results mean, sealed predictions appendix) from finished simulations. Deterministic unless generateFixes is true (then the model drafts fixes and a one-line summary per set, if enabled on the account); otherwise the recommended fixes are left as placeholders for the analyst. Returns the markdown plus claim-lint warnings to resolve before sending. Seals each set's prediction as a side effect. Results are simulated, not real-user data. For a branded PDF, open the same report in the app at /dashboard/ad-lab/pretest and use Download PDF.

Input parameters:

- `generateFixes` (boolean): Ask the model to draft 3-5 concrete fixes and a one-line summary per set. Default false.
- `language` (string): Report language. Defaults to en.
- `register` (string): Writing register. Defaults to neutral.
- `regulated` (boolean): True for regulated sectors; claims are linted harder.
- `sets` (array, required): 1-12 finished runs, one report section each.
- `title` (string, required): Report title.

Output parameters:

- `fixes` (array)
- `fixesEnabled` (boolean)
- `markdown` (string)
- `sealedPredictions` (array)
- `sets` (array)
- `source` (string)
- `warnings` (array)

### `aetherya_compare_creatives` (~335 tokens)

Compare image creatives

Compare 2+ image creatives (thumbnails, display banners, social images) against a synthetic audience. Creates the experiment, uploads the images, attaches the audience and drafts the comparison (free). It then returns status 'needs_price_confirmation' with the quoted simulation sessions; nothing is spent until you re-call with accept_sessions >= the quote (pass accept_sessions up front to skip the pause). To confirm a draft already created, pass comparison_id + accept_sessions only. Cost scales with creatives x personas; an audience group is capped at 10 personas per segment. choice_task adds a forced-choice round (choiceCost in the quote). Collect the ranking with aetherya_get_result kind 'creative_comparison'. Results are simulated, not real-user data.

Input parameters:

- `accept_sessions` (number): Sessions you accept to spend; must cover the quote.
- `audience_group_id` (string): Audience group (see aetherya_list_audiences).
- `choice_task` (boolean): Also run a forced-choice task (each persona picks one creative). Adds sessions; the quote shows choiceCost.
- `comparison_id` (string): Confirm and run a draft created by an earlier call.
- `creatives` (array): 2-10 creatives to compare. Needed unless you pass comparison_id.
- `exposure` (string): glance = a split-second scroll-past view.
- `name` (string): Name for the comparison.
- `persona_ids` (array): Persona ids (see aetherya_list_audiences), instead of an audience group.
- `platform` (string): Where the creatives run.

Output parameters:

- `choiceCost` (number)
- `comparisonId` (string)
- `estimatedSessions` (number)
- `experimentId` (string)
- `next` (string)
- `source` (string)
- `status` (string)

### `aetherya_pretest_video` (~466 tokens)

Pretest a video

Pretest a short video (mp4, mov, webm; max 512 MB, 10 minutes) against a synthetic audience: hook score, attention curve, drop-off points. Uploads the file (video_path on this machine, or video_url), splits it into scenes, then returns status 'needs_price_confirmation' with the quote. The quote exists only after preparing, and the draft is free EXCEPT speech-to-text: when you pass no transcript_srt, preparing charges 0.25 to 6 simulation sessions for transcription up front. Analysis cost is personas x 5 + scenes x 0.5 sessions, minimum 10, max 60 personas. Nothing more is spent until you re-call with accept_sessions >= the quote (pass it up front to skip the pause). If preparing is still running the tool returns 'preparing' with experiment_id, video_id and prepare_job_id: call again with those. Collect with aetherya_get_result kind 'video_analysis'. Results are simulated, not real-user data.

Input parameters:

- `accept_sessions` (number): Sessions you accept to spend; must cover the quote.
- `audience_group_id` (string): Audience group (see aetherya_list_audiences).
- `experiment_id` (string): Resume: from an earlier call.
- `hook_windows` (boolean): Split scenes into hook windows so the opening seconds are scored on their own. Default true.
- `name` (string): Name for the pretest.
- `persona_ids` (array): Persona ids, instead of an audience group.
- `placement` (string): Where the video runs.
- `prepare_job_id` (string): Resume: from an earlier call.
- `segments` (array): Limit an audience group to these segment ids.
- `transcribe_language` (string): e.g. ro or en-US.
- `transcript_srt` (string): SRT transcript; skips the speech-to-text charge.
- `video_id` (string): Resume: from an earlier call.
- `video_path` (string): Local mp4/mov/webm file (disabled when AETHERYA_MCP_ALLOW_LOCAL_FILES=0).
- `video_url` (string): http(s) link to a video file.

Output parameters:

- `estimatedSessions` (number)
- `experimentId` (string)
- `next` (string)
- `personaCount` (number)
- `prepareJobId` (string)
- `sceneCount` (number)
- `source` (string)
- `status` (string)
- `videoId` (string)

### `aetherya_seal_prediction` (~142 tokens)

Seal a prediction before the outcome is known

Lock a prediction (a creative comparison ranking, a video analysis, a study or a manual payload) with a SHA-256 hash and timestamp, so it can later be compared with what really happened. Pass source_id to seal a finished simulation, or payload for a manual prediction. Returns the id, hash and sealed payload.

Input parameters:

- `label` (string, required): Short name, e.g. 'Spring campaign, 3 creatives'.
- `payload` (object): Prediction content; required for 'manual'.
- `source_id` (string): Id of the comparison, video analysis or study run.
- `source_kind` (string, required): What is being sealed.

Output parameters:

- `createdAt` (string)
- `id` (string)
- `payload`
- `payloadSha256` (string)
- `source` (string)

### `aetherya_verify_prediction` (~41 tokens)

Verify a sealed prediction

Recompute the hash of a sealed prediction and report whether it still matches the stored one.

Input parameters:

- `id` (string, required): Sealed prediction id.

Output parameters:

- `id` (string)
- `recomputedHash` (string)
- `source` (string)
- `storedHash` (string)
- `valid` (boolean)

### `aetherya_record_outcome` (~161 tokens)

Record the real outcome of a sealed prediction

Attach what actually happened to a sealed prediction (append-only) and get the evaluation. creative_comparison actual: {metric:'ctr', values:[{creativeId?|variant?, value, impressions?, clicks?}]}; impressions and clicks let Aetherya record the result for calibration (see calibrationRecorded). video_analysis actual: {threeSecondHoldPct?, biggestDropSec?, watchedFullPct?}. Other kinds: any object.

Input parameters:

- `actual` (object, required): The observed outcome.
- `id` (string, required): Sealed prediction id.
- `note` (string): Optional free-text note stored with the outcome.
- `source` (string): Where the numbers came from, e.g. 'client_email_2026-10-09'.

Output parameters:

- `calibrationRecorded`
- `evaluation`
- `id` (string)
- `source` (string)

### `aetherya_create_intake_link` (~166 tokens)

Create a client file-intake link

Create a secret upload link to send a client so they can drop creative files (images, video) for pretesting. Returns the url and expiry; treat the url as a secret and share it only with the client. Needs a workspace admin or manager, and the feature must be enabled for the workspace (otherwise a clear 'not enabled' message).

Input parameters:

- `expires_in_days` (integer): Days until the link expires (1-30, default 14).
- `label` (string, required): Who or what the link is for, e.g. 'Acme spring banners'.
- `max_files` (integer): Max files (default and cap 50).
- `max_total_mb` (integer): Max total size in MB (default and cap 2048).

Output parameters:

- `expiresAt` (string)
- `id` (string)
- `source` (string)
- `url` (string)

### `aetherya_list_intake_files` (~71 tokens)

List files uploaded through an intake link

List the files a client uploaded through an intake link (name, type, size, status, set label, a short-lived download url). The feature must be enabled for the workspace.

Input parameters:

- `link_id` (string, required): Intake link id from aetherya_create_intake_link.

Output parameters:

- `files` (array)
- `source` (string)

### `aetherya_list_audiences` (~108 tokens)

List your synthetic audiences

List the synthetic audiences (simulated personas) in this workspace: their overall shape (where they are, what they do, how much is actually filled in) plus a sample with ids. These are simulated personas, not real people, and a stored persona pool is usually narrower than the audience a decision is about — read the distribution before targeting by id, and prefer describing the audience you want in aetherya_plan_study when the pool does not match it.

Output parameters:

- `audiences` (array)
- `count` (number)
- `source` (string)
- `total` (number)

### `aetherya_create_audience` (~360 tokens)

Create a synthetic audience

Create synthetic-audience personas to simulate against. Describe who you want to model and get back personas (with ids) you can then target as the 'audience' in aetherya_audit_site, aetherya_test_copy, or aetherya_ask_audience. These are SIMULATED personas (engine-generated), not real people. Returns the created personas inline.

Input parameters:

- `brand_context` (object): Optional. The brand under test and its competitors, so personas carry brand awareness for that market. Requires segments or website_url; otherwise it is rejected.
- `category` (string): Optional. A category/segment label for the audience (e.g. 'e-commerce', 'fintech').
- `count` (integer): How many synthetic personas to create (1-25, default 3).
- `description` (string): Who to simulate, in plain language (e.g. 'price-sensitive first-time SaaS buyers in Germany' or 'busy parents shopping for groceries online'). When provided, Aetherya engine-generates personas matchi…
- `language` (string): Optional. Language for the personas, e.g. 'ro'.
- `market` (string): Optional. Target market / locale as a country or region (e.g. 'US', 'DE', 'UK').
- `pack_name` (string): Optional. Name for the saved audience pack.
- `segments` (array): Optional. Explicit audience segments (max 12); weights follow 'size'. Builds a manual blueprint.
- `website_url` (string): Optional. A site to infer the audience from — Aetherya analyzes it and generates personas that match its likely visitors.

Output parameters:

- `audiences` (array)
- `count` (number)
- `source` (string): Always 'synthetic_audience_simulation'.

## Diagnostics

Captured diagnostic sections: Provenance, Vulnerabilities, Dependencies. The full working is on the page: https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp#diagnostics

## Score history

- 2026-11-05: 68
- 2026-11-04: 67
- 2026-11-03: 67
- 2026-11-02: 66
- 2026-11-01: 66
- 2026-10-31: 65
- 2026-10-30: 65
- 2026-10-29: 64
- 2026-10-28: 64
- 2026-10-27: 63
- 2026-10-26: 63
- 2026-10-25: 62
- 2026-10-24: 62
- 2026-10-23: 67
- 2026-10-22: 66
- 2026-10-21: 66
- 2026-10-20: 65
- 2026-10-19: 54

## Common questions

### What is the Aetherya MCP server?

Aetherya is an MCP server listed in the public MCP registry as ai.aetherya/mcp. Test decisions on a synthetic audience: studies, site audits, copy tests and post simulations. This page covers its npm package (@aetherya/mcp).

### Is the Aetherya MCP server safe to use?

Aetherya scores 68 out of 100 on VerifyMCP. We recorded 1 known advisory against it as of 5 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Aetherya MCP server expose?

Aetherya exposes 27 tools: aetherya_list_capabilities, aetherya_estimate, aetherya_plan_study, aetherya_confirm_study, aetherya_read, and 22 more. Their descriptions and schemas cost roughly 6,901 tokens of context every time the server is loaded.

### Is the Aetherya MCP server still maintained?

Aetherya is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Aetherya MCP server under?

Aetherya declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/@aetherya/mcp
- Socket report: https://socket.dev/npm/package/@aetherya/mcp
- Website: https://aetherya.ai/
- Changelog RSS feed: https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp.json
- HTML version of this page: https://verifymcp.io/servers/ai-aetherya-mcp/aetherya-mcp