For AI agents — MCPBundles Directory API
For MCP client apps (human connect):
MCP connect URL: https://mcp.mcpbundles.com/bundle/yutori
Skill page: https://www.mcpbundles.com/skills/yutori
Tools: 15
Live probe refreshed Nov 21, 2026 · Endpoint host mcp.mcpbundles.com
AI Skill
SKILL.md
Domain knowledge for Yutori — workflow patterns, data models, and gotchas for your AI agent.
Yutori
Yutori builds reliable web agents that autonomously execute tasks on the web. Offers four APIs: n1 (pixels-to-actions LLM), Browsing (cloud browser automation), Research (deep web research with 100+ MCP tools), and Scouting (continuous web monitoring with configurable schedules and alerts).
Data Model
Scouts are continuous web monitoring jobs. Each scout has an ID, configurable schedule, and produces updates.
Scout Updates are the output of a scout run — changes detected, snapshots, or alerts.
Usage and Health are account-level endpoints for validation and quota checks.
Gotchas
API coverage: Browsing (one-shot cloud browser tasks), Research (one-shot deep web research), and Scouting (continuous monitoring) are all exposed here, plus health and usage reads. The n1 pixels-to-actions API is not exposed in this bundle.
Async tasks: Browsing and research return a task_id — poll task status until status is succeeded or failed.
Scouts: List scouts for IDs, then fetch scout details or paginated updates for findings over time.
Pagination: Use page_size when listing scouts and updates to control response size.
Explore capabilities (15)
read 7
write 8
Yutori
15
Yutori Create Browsing Task
write
yutori_create_browsing_task
Create a browser automation task on Yutori. An AI agent opens a cloud browser at the specified URL and performs the described task — clicking, typing, scrolling, and navigating like a human. Use for: filling forms, extracting data from web pages, checking for changes, logging into websites, downloading information. Returns a task_id for tracking status and a view_url to watch the agent in real-time. Poll browsing task status with the returned task_id.
MCPBundles publishes this directory for MCP discovery. Except where we host or operate an offering, third-party services run under their own terms. Product and company names on this page are used in a descriptive, identifying way (including under nominative fair use where applicable); they remain the property of their owners. Nothing here grants you rights in those marks, and nothing here is an offer to sell a third party's services. Terms
Yutori
Add to your AI client, then sign in once when you connect.
Yutori builds reliable web agents that autonomously execute tasks on the web. Offers four APIs: n1 (pixels-to-actions LLM), Browsing (cloud browser automation), Research (deep web research with 100+ MCP tools), and Scouting (continuous web monitoring with configurable schedules and alerts).
This listing identifies a third-party integration. MCPBundles is not affiliated with, sponsored by, or endorsed by the service or its operator unless stated on this page.
Yutori Create Browsing Task — Create a browser automation task on Yutori. An AI agent opens a cloud browser at the specified URL and performs the described task — clicking, typing, scrolling, and navigating like a human. Use for: filling forms, extracting data from web pages, checking for changes, logging into websites, downloading information. Returns a task_id for tracking status and a view_url to watch the agent in real-time. Poll browsing task status with the returned task_id.
Yutori Create Research Task — Create a deep web research task on Yutori. Unlike browsing (single browser agent), research deploys 100+ MCP tools for full investigation across the web. Use for: market research, competitive analysis, technology landscape surveys, news summarization, trend analysis, or any query requiring wide and deep web research. Returns a task_id to track progress. Poll research task status with that task_id. Results include citations with source URLs.
Yutori Create Scout — Create a new Yutori scout for continuous web monitoring. Scouts periodically research the web for changes matching your query and send alerts via email and/or webhooks. Unlike browsing (one-time browser task) or research (one-time deep research), scouts run on a recurring schedule. Use for: tracking competitor launches, monitoring market changes, watching for regulatory updates, tracking mentions or news about specific topics. Returns the scout ID, schedule details, and view URL.
Yutori Delete Scout — Permanently delete a Yutori scout and all its historical updates. This action cannot be undone. The scout stops monitoring and all data is removed. Prefer marking a scout done if you want to stop monitoring but preserve findings.
Yutori Get Browsing Task — Get the current status and results of a Yutori browsing task. Statuses: 'queued' (waiting to start), 'running' (agent is actively browsing), 'succeeded' (task completed successfully — check 'result' and 'structured_result'), 'failed' (task failed). Poll this endpoint to wait for task completion. Returns the task status, text result, structured result (if output_schema was provided), and a view_url to watch the agent.
Frequently Asked Questions
What is the Yutori MCP server?
Yutori builds reliable web agents that autonomously execute tasks on the web. Offers four APIs: n1 (pixels-to-actions LLM), Browsing (cloud browser automation), Research (deep web research with 100+ MCP tools), and Scouting (continuous web monitoring with configurable schedules and alerts). It provides 15 tools that AI agents can use through the Model Context Protocol (MCP).
How do I connect Yutori to my AI agent?
Add the MCPBundles server URL to your MCP client configuration (Claude Desktop, Cursor, VS Code, etc.). The URL format is: https://mcp.mcpbundles.com/bundle/yutori. Authentication is handled automatically.
Yutori Get Research Task — Get the current status and results of a Yutori research task. Statuses: 'queued' (waiting to start), 'running' (research in progress), 'succeeded' (completed — check 'result', 'structured_result', and 'updates' with citations), 'failed' (task failed). Poll this endpoint to wait for completion. Results include the research text, structured data (if output_schema was provided), update history with citations (URLs and preview data), and stats (tool calls, websites visited, time saved).
Yutori Get Scout — Get full details of a specific Yutori scout including its query, status, schedule, timezone, next run time, update count, and creation/completion timestamps. Statuses: 'active' (running on schedule), 'paused' (stopped temporarily), 'done' (completed). Use scout updates to see what this scout has found.
Yutori Get Scout Updates — Get the updates (findings) from a Yutori scout. Each update contains the research content found during a monitoring run, citations with source URLs and preview data, stats (tool calls, websites visited, time saved), and optional structured results. Updates are ordered by recency. Use this to see what a scout has discovered over time.
Yutori Get Usage — Get your Yutori API usage statistics including the number of active scouts and their IDs. Use to monitor your current resource consumption.
Yutori Health Check — Check the health and connectivity of the Yutori API. Returns the API status. Use to confirm credentials are valid and the API is reachable.
Yutori List Scouts — List all Yutori scouts (continuous web monitoring tasks). Scouts periodically research the web for changes matching a query and send alerts. Returns scout IDs, queries, statuses, display names, schedules, and pagination info. Filter by status (active/paused/done). Use scout detail and scout-updates reads for a specific scout.
Yutori Mark Scout Done — Mark a Yutori scout as done, stopping its continuous monitoring. The scout's status changes to 'done' and it will no longer check for updates. Historical updates are preserved and can still be retrieved. You can restart monitoring later with the restart-scout action.
Yutori Restart Scout — Restart a paused or completed Yutori scout. The scout resumes continuous monitoring with its existing query, schedule, and webhook configuration. Its status changes back to 'active'.
Yutori Update Scout — Partially update a Yutori scout's configuration. Only provided fields are changed; omitted fields remain unchanged. Use to modify the query, change the monitoring interval, update webhook settings, or toggle public visibility.
Yutori Update Scout Email Settings — Update the email notification settings for a Yutori scout. Toggle whether the scout sends email alerts when it finds new updates.
Yutori provides 15 tools that can be called by AI agents, along with a SKILL.md that gives your AI agent domain knowledge about when and how to use them.
What authentication does Yutori require?
Yutori uses API Key. Yutori requires credentials. Connect via MCPBundles and authentication is handled automatically.
MCPBundles is an independent platform built on the open Model Context Protocol standard. MCPBundles is not affiliated with, sponsored by, or endorsed by Anthropic PBC.
More in machine learning
Other MCP servers in this category from the directory index