Intelogos MCP

Intelogos MCP is a Model Context Protocol (MCP) server that connects an AI assistant you already use to your organization's Intelogos data. Once connected, the assistant can look up people, departments and projects, pull KPIs, read daily activity and work sessions, analyze application usage and check schedules and time off, then answer your questions in plain language.

It is a remote server run by Intelogos. You don't install or host anything: you add the server URL and your organization's credentials to the client.

Intelogos MCP is included in the AI Intelligence and Enterprise plans. See the pricing page for what each plan includes.

---

Who Can Connect

Only an Account Owner or Admin can create the connection, from Settings → MCP. That tab issues a Client ID and Client Secret for your organization; managers and regular users don't see it. See the Settings guide for the tab itself.

Your MCP client signs in with those credentials over OAuth 2.0. Requests without a valid token are refused, so the server URL alone gives no access to any data.

---

Set Up Your Client

1. In Intelogos, open Settings → MCP and copy the MCP URL, Client ID and Client Secret from the MCP Connection card.
2. In your AI client, open its MCP settings (called Connectors, MCP servers or Integrations depending on the client) and add a new remote server.
3. Enter the MCP URL as the server URL and the Client ID and Client Secret as its credentials, then save.
4. Restart the client if it asks you to. The Intelogos tools appear in the assistant's tool list once the connection is active.

The MCP tab shows the same steps with tabs for General, ChatGPT and Claude configurations.

Claude (claude.ai)

Open Settings → Connectors, choose Add custom connector and paste the MCP URL as the server URL. Enter the Client ID and Client Secret where the connector asks for the server's credentials, then save.

Claude Code

Add the server from the terminal, using the MCP URL from Settings → MCP:

claude mcp add --transport http intelogos <MCP URL from Settings → MCP>

Cursor

Add the server to .cursor/mcp.json in a project, or to ~/.cursor/mcp.json for every project, using the MCP URL from Settings → MCP:

{
  "mcpServers": {
    "intelogos": {
      "url": "<MCP URL from Settings → MCP>"
    }
  }
}

Claude Code and Cursor then need the Client ID and Client Secret from Settings → MCP; enter them where the client asks for the server's credentials.

ChatGPT and other clients

Add a remote MCP server with the same MCP URL and credentials in the client's MCP or connector settings. Menu names differ by client and version; the four generic steps above apply everywhere.

---

What Your Assistant Can Access

The server exposes 39 tools, listed here by purpose as a connected client reported them on 2026-09-30.

Search and lookup

Find a person, department or project, then fetch its record.

Tool	What it does
search	Search employees, departments and projects
fetch	Retrieve a full document by ID

People, teams and org structure

Who is in the organization and how it is organized.

Tool	What it does
get_employee_list	Paginated list of employees with filtering and sorting
get_departments_list	Departments with structure and employee counts
get_projects_list	Projects and teams with members and status
get_managers_hierarchy	Manager and subordinate reporting hierarchy
get_org_helper_data	Organization helper data for filtering
get_organization_settings	Working hours, week start and other org settings

Performance and KPIs

KPIs, rankings and highlights for people, departments and groups.

Tool	What it does
get_organization_users_data	Performance metrics with flexible grouping
get_key_performance_stats	KPI aggregates and time series for users or groups
get_leaderboard	Ranked performance data
get_user_highlights	Performance highlights and top tools for one user
get_department_spotlight	Performance analytics for departments
get_department_workload_analysis	Workload metrics grouped by department

Daily activity and work sessions

How a day, a session or an hour was spent.

Tool	What it does
get_daily_chronicle_summary	One synthesized row per employee for a date
get_daily_work_sessions	Daily work-session summary for a date
get_employee_work_sessions	Work session timeline per user
get_employee_daily_info_report	Daily breakdown of apps, windows and projects for a user
get_detailed_info_report_grouped	Activity grouped by app, window or project
get_detailed_info_report_drill_down	Row-level detail for one app, window or project
get_detailed_info_report_light	Raw chronological activity for a user
get_hourly_activity	Per-hour activity for one employee on one day

Work time, schedules and policies

Worked time against expected hours and the configured schedule.

Tool	What it does
get_employees_work_time_report	Work time analytics
get_employees_policy_work_time	Expected hours and schedule per employee
get_work_policies	The organization's configured work schedule
get_availability_progress	Worked time against required time

Applications and tools

Which applications and websites are used, and how they are categorized.

Tool	What it does
get_users_application_usage	Application usage per user
get_tools_usage	Applications and tools used in a date range
get_time_categories	Usage aggregated into the organization's categories
get_application_category_usage	Usage by category (primary, secondary, distracting)
get_engagement_category_usage	Use of high, medium and low engagement tools
get_tool_category_management	Application categorization rules
get_tool_importance	Importance rules for a named app or website
get_tool_types	Assignable tool types (work categories)
get_assigned_smart_projects	Smart project auto-tracking rules

Time off

Time-off requests, balances, policies and the team calendar.

Tool	What it does
get_time_off_calendar	Time-off calendar across the team
get_time_off_organization_requests	Time-off requests with filters
get_time_off_policies	Time-off policies
get_time_off_user_balance	Remaining time-off balance for a user

---

Example Prompts

Try these first. The answers are illustrative and use fictional teams and numbers, not data from a real organization.

"Which departments had the lowest availability last week?" Example answer with sample data: Across 4 departments, Customer Support had the lowest availability at 81% of expected hours, followed by Design at 88%. Engineering and Sales were both above 92%.

"Summarize how the design team spent its time yesterday." Example answer with sample data: The 6-person Design team logged 41 active hours. 62% went to design tools, 21% to communication apps and 9% to browsers. Two people had more than 90 minutes of idle time in the afternoon.

"Who is on time off next week, and will any team be short-staffed?" Example answer with sample data: Three approved requests overlap next week: two in Sales (Monday to Wednesday) and one in Engineering (all week). Sales will be at 60% headcount on Monday and Tuesday.

"Show the top tools used by the Sales department this month and flag any distracting ones." Example answer with sample data: Top tools: the CRM (38% of active time), email (22%) and video calls (15%). Two tools are categorized as distracting: a video streaming site (2.1 hours in total) and a news site (1.4 hours).

"Compare this week's productivity KPI with last week for the whole organization." Example answer with sample data: Productivity is 84% this week, up 3 points from 81% last week. Activity is flat at 77%. The largest gain came from Engineering (up 6 points); Customer Support fell 2 points.

---

Managing Access

Admins keep control of the connection from Settings → MCP:

• Regenerate issues a new Client Secret and invalidates the previous one. Update every connected client afterwards.
• Deactivate pauses AI access without deleting the connection.
• Delete removes the connection permanently.

---

Related

• Ask AI: the built-in assistant inside Intelogos.
• Managing with AI: the questions managers ask most.
• Intelogos MCP overview and pricing.