MCP Integration

UserTold captures consented in-product interviews: voice and transcripts, in-page behavior and page context, plus screen sharing when a participant approves it in a supported desktop browser.

Findings become source-linked Evidence traceable to what participants said, did, and saw, then draft Work for review. Generated transcripts and analysis do not replace the captured source.

UserTold does not recruit participants or replace consent. Mobile and unsupported browsers continue without screen sharing. Denied permissions, interruptions, and failures can leave capture gaps.

Fast path: let your agent work with existing research

Help me connect UserTold to this agent with https://mcp.usertold.ai/mcp and complete browser authorization.

Read usertold://organizations and usertold://projects. Ask me to select or create a Project, then proceed. Propose changes before acting.

Connect

Server:

https://mcp.usertold.ai/mcp

Clients use OAuth 2.1 browser authorization with PKCE; no API token is needed.

Discover the current surface

Discover all 19 tools with tools/list. Two bounded organization/project resources expose references. Get the Project's install-once snippet, run the preflight, and resume with studies.list. There is no hidden tool or prompt catalog.

Raw voice/screen playback and private administration stay in the dashboard.

Product boundaries

  • Activating changes capture and needs approval. In studies.update, active starts matching interviews; closed stops new ones without interrupting one in progress.
  • studies.create/update use CLI invitation (reward/direct_link) and bounded allowedOrigins; null uses built-in, [] clears, presets fail. Get returns public recruitmentUrl, no identity/claims/settings. projects.get_widget_setup is sole snippet source.
  • Review source evidence and interview context before readying Work.
  • Retry durable failures or 24-hour stalls only; inspect interviews.processing_status. Never auto-retry. Manual Evidence stays.
  • work.push sends ready Work to GitHub or Linear; it does not decide readiness.

If setup fails

Reconnect authorization. For capture failures, run the preflight, then confirm browser execution and permissions in the dashboard.

Continue