Skip to main content
Glama
kubecon-eu-mcp

kubecon-eu-mcp

by njoerd114

kubecon-eu-mcp

Python 3.11+ License MIT KubeCon EU 2026

An MCP server for KubeCon + CloudNativeCon Europe 2026. This tool serves as an agentic schedule planner, party optimizer, and conference guide for the event held March 23-26 at RAI Amsterdam.

Overview

kubecon-eu-mcp connects AI agents to live conference data, enabling them to help attendees navigate the schedule, discover social events, and plan their trip. It fetches data live from upstream sources with efficient in-memory caching to ensure accuracy throughout the week.

Data sources include:

As a meta tie-in, this project highlights the Agentics Day: MCP + Agents co-located event on Monday, March 23.

Related MCP server: agent-zone-mcp

Features

  • Smart Scheduling: Search talks by topic, speaker, or technology (eBPF, AI, WASM, etc.).

  • Party Optimizer: Find evening social events and plan the best route through Amsterdam.

  • Logistics Guide: Access hotel block info, airline discounts, and venue transit details.

  • Agentic Planning: Built-in prompts to help users build 4-day itineraries or first-timer guides.

  • Session Scoring: Personalized session ranking using the scoring rubric from kubecon-event-scorer by Fredrik Carlsson — scores across role relevance, topic alignment, and strategic value.

  • Live Updates: Data is pulled directly from official sources to reflect last-minute room changes.

Quick Start

1. Zero Install (uvx)

If you have uv installed, you can run the server without manual installation:

uvx kubecon-eu-mcp

2. Standard Install (pip)

Install the package from your local clone or repository:

pip install .
# Then run the server
kubecon-eu-mcp

3. Docker

Build and run the containerized server:

docker build -t kubecon-eu-mcp .
docker run -it kubecon-eu-mcp

Usage Examples

Once connected to an MCP client like Claude Desktop, you can ask questions like:

  • "What AI talks are on Wednesday?"

  • "Find all sessions featuring Lin Sun."

  • "What's the best way to get to RAI Amsterdam from Schiphol airport?"

  • "Are there any parties near the venue on Tuesday night?"

  • "Plan a 4-day itinerary for a platform engineer interested in security."

Example Interaction

User: What AI talks are on Tuesday? Assistant: (Calls search_sessions(query="AI", day="tuesday")) The agent searches the live schedule and returns matching sessions with titles, speakers, rooms, and times. It can then help you compare options, flag conflicts with other sessions you want to attend, and suggest alternatives.

Tools Reference

The server exposes 12 specialized tools:

Tool

Description

search_sessions

Search talks by keyword, topic, or speaker.

get_schedule

Get the full schedule for a specific day.

find_speaker

Look up sessions for a specific person.

find_parties

Discover social events and happy hours.

plan_party_route

Get an optimized route for evening events.

get_venue_info

Details on rooms, maps, and venue address.

get_hotel_info

Hotel block rates and distances to RAI.

get_travel_info

Airport, transit, and airline discount codes.

get_colocated_events

Monday's specialized events (ArgoCon, Agentics Day, etc.).

get_conference_overview

High-level event summary and key dates.

score_sessions

Get sessions with a personalized scoring rubric for AI-powered ranking.

detect_conflicts

Check if selected sessions overlap in time.

Resources Reference

Access structured data directly via these URIs:

  • kubecon://overview: Full conference summary.

  • kubecon://venue: Venue layout and transit details.

  • kubecon://hotels: Accommodation options.

  • kubecon://colocated-events: Monday's co-located event list.

Prompts Reference

Pre-configured workflows for common tasks:

  • plan_my_kubecon: Builds a personalized 4-day itinerary based on interests.

  • party_tonight: Plans an evening of networking and social events.

  • first_timer_guide: Provides essential tips for new attendees.

  • whats_happening_now: Finds sessions starting soon based on current time.

  • create_profile: Interactive profile builder that scores and ranks sessions using the kubecon-event-scorer rubric.

Configuration

Claude Desktop

Add this to your claude_desktop_config.json:

Standard (stdio) Mode:

{
  "mcpServers": {
    "kubecon-eu-mcp": {
      "command": "uvx",
      "args": ["kubecon-eu-mcp"]
    }
  }
}

Hosted (HTTP) Mode: If running the server with the --http flag:

{
  "mcpServers": {
    "kubecon-eu-mcp": {
      "url": "http://localhost:8000/mcp"
    }
  }
}

Contributing

  1. Clone the repository.

  2. Install dependencies with uv sync or pip install -e ..

  3. Run the server in development mode: python -m kubecon_eu_mcp.

  4. Submit a Pull Request with your improvements.

License: MIT

Available Tools

12 tools
detect_conflictsA

Detect scheduling conflicts among selected sessions.

Checks whether any of the provided sessions overlap in time, helping attendees resolve conflicts in their planned schedule.

Inspired by kubecon-event-scorer's conflict detection.

Args: session_uids: Comma-separated session UIDs to check for conflicts.

Returns: JSON with conflict pairs, overlap duration, and session details.

Inputs: session_uids (required). Returns: result

find_partiesA

Find conference parties, happy hours, and social events.

Args: day: Optional day filter: "monday", "tuesday", "wednesday", "thursday". after: Optional time filter — only show events starting after this time (e.g., "6PM"). before: Optional time filter — only show events starting before this time (e.g., "10PM").

Returns: JSON array of parties with name, time, sponsor, location, and RSVP link.

Inputs: day, after, before. Returns: result

find_speakerA

Find sessions by a specific speaker.

Args: name: Speaker name or partial name (e.g., "Lin Sun", "Bryce").

Returns: JSON array of sessions featuring the speaker.

Inputs: name (required). Returns: result

get_colocated_eventsA

Get Monday's co-located events (ArgoCon, CiliumCon, Agentics Day, etc.).

Returns: JSON array of co-located events with name, duration, room, and requirements.

Inputs: no named parameters. Returns: result

get_conference_overviewA

Get a high-level overview of the entire conference.

Returns: JSON with event name, dates, schedule at a glance, and useful links.

Inputs: no named parameters. Returns: result

get_hotel_infoB

Get conference hotel block information with rates and distances.

Returns: JSON array of hotels with name, address, rate, distance to venue, and availability.

Inputs: no named parameters. Returns: result

get_scheduleA

Get the full conference schedule for a specific day.

Args: day: Day name — "monday" (co-located), "tuesday", "wednesday", or "thursday".

Returns: JSON array of all sessions for that day, sorted by start time.

Inputs: day (required). Returns: result

get_travel_infoA

Get travel information: airport, transit, parking, and airline discounts.

Returns: JSON with airport details, public transport info, parking, and airline discount codes.

Inputs: no named parameters. Returns: result

get_venue_infoA

Get venue details: address, rooms, maps, and transit information.

Returns: JSON with venue name, address, room list, map links, transit info, and parking.

Inputs: no named parameters. Returns: result

plan_party_routeA

Plan an optimized party-hopping route for an evening.

Returns parties for the requested day sorted by time, with location details to help plan an efficient route through Amsterdam.

Args: day: Day name — "monday", "tuesday", "wednesday", or "thursday". preferences: Optional preferences like "networking", "food", "near venue", "open bar".

Returns: JSON with sorted party timeline and location context for route planning.

Inputs: day (required), preferences. Returns: result

score_sessionsA

Get sessions ready for personalized scoring, with a scoring rubric.

Returns scorable sessions (logistics events filtered out) along with a structured scoring rubric. You (the AI agent) should apply the rubric to rank sessions for the attendee.

Scoring rubric is based on the kubecon-event-scorer project by Fredrik Carlsson.

Args: role: Attendee's job title (e.g., "Platform Engineer", "SRE", "Developer"). interests: Comma-separated interests (e.g., "eBPF, security, AI on Kubernetes"). day: Optional day filter: "monday", "tuesday", "wednesday", "thursday". experience_level: "beginner", "intermediate", "advanced", or "expert". priorities: Comma-separated goals (e.g., "evaluate service mesh tools, learn GitOps"). prefer_hands_on: Boost hands-on workshops and labs. prefer_deep_dives: Boost deep technical talks over intros. avoid_vendor_pitches: Penalize vendor-heavy marketing sessions. limit: Max sessions to return (default 30).

Returns: JSON with attendee profile, scoring rubric, and session list to score.

Inputs: role (required), interests (required), day, experience_level, +5 more. Returns: result

search_sessionsA

Search conference sessions by keyword, topic, speaker name, or technology.

Args: query: Search query (e.g., "eBPF", "security", "AI agents", "platform engineering"). day: Optional day filter: "monday", "tuesday", "wednesday", "thursday". track: Optional track filter (e.g., "Keynote", "Tutorial", "Breakout"). limit: Maximum number of results to return (default 20).

Returns: JSON array of matching sessions with title, speakers, time, room, and URL.

Inputs: query (required), day, track, limit. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4/5.0

    Scored across 12 tools

    Disambiguation5/5

    Each tool targets a distinct aspect of conference attendance (scheduling, parties, travel, scoring, etc.) with no overlapping purposes. Even similar tools like find_parties and plan_party_route are clearly differentiated by scope and output.

    Naming Consistency5/5

    All tool names use a consistent verb_noun pattern in snake_case (e.g., get_schedule, find_speaker, plan_party_route). Verbs are uniform ('get' for retrieval, 'find' for search, etc.) with no mixing of conventions.

    Tool Count5/5

    12 tools is well-scoped for a conference MCP server, covering essential attendee needs without overwhelming. Each tool serves a clear purpose, and the count aligns with typical domain coverage.

    Completeness4/5

    The tool set covers core conference activities (schedule, sessions, speakers, travel, parties, conflict detection, scoring). A minor gap is the lack of a dedicated speaker bio tool, but search_sessions and find_speaker provide sufficient session-level detail.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that enables AI assistants to interact with Kubernetes clusters by translating natural language into kubectl and Helm operations. It allows users to query, manage, and diagnose Kubernetes resources and cluster states through a seamless integration.
      19
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server for the AI Engineer Conference 2025, enabling talk submissions and conference information retrieval.
      57
      MIT