Startup idea discovery
MSX turns live market signals into ranked startup ideas with evidence that agents can use instead of starting from a blank search.
MSX
MSX is a market intelligence runtime and API for AI agents that helps them find startup ideas, run live market investigations, and return source-backed opportunities. It is aimed at builders and teams using Claude, Codex, Cursor, or internal agents.
Overview
MSX is a market intelligence runtime and API for AI agents. It helps agents such as Claude, Codex, and Cursor find startup ideas, decide what to build next, and run live market investigations against public demand signals.
The product is built around a hosted skill file at `msx.dev/skill.md`, a REST API, and source-backed investigation outputs. The workflow is designed so an agent can authenticate, ask before billing when needed, run a research job, and return buildable opportunities with citations and practical next actions. The enterprise pages also position MSX as a broader VOC and market-signal layer that mines web channels continuously for customer language, unmet needs, and launch or positioning inputs.
Core capabilities
Demand-signal scouting
The docs describe a workflow that can scan public demand, compare existing tools, and surface where buyers are asking for something better.
Agent authentication and API access
Agents can authenticate through a device flow, receive an account-bound API key, and then use the REST API for investigations and account checks.
Durable investigation workflow
Investigations are durable jobs with status, progress, and result endpoints, so agents can poll and resume rather than treating research as a one-shot request.
Source-backed results
Returned research includes citations, urgency, confidence, and practical next actions, which makes the output easier to use in downstream planning or drafting.
Web-wide signal mining
The enterprise page says MSX can mine 30+ web channels continuously for VOC, idea validation, talent signals, and market movement.
Common use cases
Validating startup ideas
A founder or operator can ask an agent to scout a broad market question, then use MSX to rank opportunities and review source-backed summaries before deciding what to build.
Market and VOC research
A PMM or research team can use the enterprise workflow to gather customer language, segment-tagged proof, and citation-grade evidence for positioning or launch planning.
Agent-native research workflows
A builder can connect an agent to the skill file and API so market research runs inside the same workflow where they already work, rather than moving into a separate portal.
Ongoing market monitoring
A team can run recurring investigations as durable jobs, then compare saved results and source-backed reports over time to track market movement.
Enterprise research support
An enterprise buyer can contact sales for managed research, custom PDF or markdown outputs, or agent-ready market intelligence workflows.
Pros and Cons
Pros
- Built specifically for agents, with a hosted skill file and API flow instead of a manual research workflow.
- Produces source-backed investigations with citations, urgency, confidence, and next actions.
- Supports durable investigation jobs, which is useful when research takes more than a single request.
- Offers clear support paths for private account issues and setup questions.
- Has published pricing tiers with included API access and saved results.
Cons
- The documentation is stronger than the capability breakdown, so some workflow details are only described at a high level.
- Pricing and support information are present, but integration specifics and data-source coverage are still limited in the public pages.
FAQ
How do agents start using MSX?
MSX is designed to give agents a simple public entrypoint: the hosted skill file at `msx.dev/skill.md` plus the REST API. The docs say agents should read the skill file before calling MSX, then follow the hosted contract for authentication, billing consent, quota handling, and investigation behavior.
How does authentication work?
The docs describe API-driven device authentication. An agent starts the device flow, shows the user a verification URL, waits for Google sign-in, and then polls for an account-bound API key before making authenticated requests.
What does an investigation return?
MSX supports durable investigation jobs. The create call returns an investigation ID and status, and the agent polls until the job succeeds, fails, cancels, or expires. When it succeeds, the result endpoint returns opportunities with evidence and next actions.
Is MSX sold as a subscription?
The pricing page lists Scout, Builder, and Growth plans. It also states that every plan includes the MSX skill, API access, saved results, and source-backed investigation reports.
How do users contact support?
Support is split by channel: email is recommended for billing, account access, privacy, deletion, or other private account details, while Discord is recommended for setup questions, product help, and agent troubleshooting.
Quick Facts
- Category
- AI agent market intelligence
- Primary users
- Agents and teams using Claude, Codex, Cursor, or internal agents
- Access model
- Hosted skill file plus REST API
- Pricing
- Paid plans listed: Scout, Builder, and Growth
- Support channels
- Email and Discord
- Domain
- mothershipx.dev / msx.dev
MSX 대안
Hype
Hype is a web tool for finding trending YouTube topics by category, time range, and scoring mode. It helps creators spot emerging ideas, inspect source videos, and decide what to cover next.
Paperpal
Paperpal 是一款面向学术写作的 AI 工具，提供文献阅读、语言润色、引用、查重、AI 检测和投稿前检查。它适合需要在英文论文和学术材料上进行写作、修改与合规自检的学生、研究人员和教师。
VForms
VForms lets you place questionnaires directly inside YouTube videos so viewers answer in context instead of switching to a separate form. It is aimed at collecting feedback and insights through timestamped, embeddable video forms.
Scite
Scite는 논문 검색, 인용 문맥 확인, 학술 문헌에 근거한 답변을 제공하는 AI 연구 플랫폼입니다. 연구자, 학생, 팀의 근거 평가를 돕습니다.
Shengsuanyun
Shengsuanyun은 효율적인 데이터 관리 및 분석을 위한 클라우드 기반 솔루션을 제공합니다.
Model Council
Perplexity Model Council is a multi-model research feature that runs a query across several AI models and combines the results into one answer. It is available to Perplexity Max subscribers on the web, with mobile app support coming soon.