Livesession
LiveSession is a session replay and user behavior analytics tool that helps you understand how users interact with your website.
Verdict
Common use cases
- Triage support tickets with live session context
- Review active users during feature rollouts
- Debug customer issues in real time
- Track session volume by time or segment
- Audit user activity for compliance checks
Integration
- Vendor
- Livesession
- Category
- developer-tools
- Auth
- API_KEY
- Tools
- 1
- Composio slug
livesession
Tools
- List Live Sessions
Tool to list live sessions. use when you need paginated session data filtered by criteria after authentication.
Setup
Setup guide
- 11. In Switchy, open your workspace settings and navigate to the Integrations tab. 2. Search for Livesession and click Connect. 3. You'll be prompted to enter your Livesession API key — generate one by logging into your Livesession dashboard, going to Settings > API, and creating a new key with read permissions for sessions. 4. Paste the key into Switchy and click Authorize. 5. To test the connection, create or open a Space and type '@Livesession list sessions from the last 24 hours' — if the MCP returns a list of active or recent sessions, you're connected. 6. From any Space, @mention Livesession followed by your query criteria to invoke the tool.
What teammates see: by default, memories from Livesession are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Recent Active Sessions
@Livesession list all sessions from the last 60 minutes. Show me user IDs and session start times.Open in a Space →
Sessions by User Segment
@Livesession list sessions from users tagged as 'enterprise' in the last 7 days. Include session duration.Open in a Space →
High-Activity Days
@Livesession list sessions from the past 30 days, grouped by date. Highlight days with more than 500 sessions.Open in a Space →
Support Ticket Context
@Livesession list all sessions for user ID 'abc123' from the last 14 days. Show session metadata and timestamps.Open in a Space →
Feature Adoption Check
@Livesession list sessions from the last 3 days where users visited '/new-feature'. Include session count and user IDs.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Livesession show me the last 10 active user sessions from today, filtered by sessions longer than 5 minutes
Retrieved 10 sessions from today with duration >5 minutes. Example entries: Session #a7f3 (user ID 8821, 12m 34s, 47 page views, last active 8 minutes ago on /checkout). Session #b2e9 (user ID 7453, 9m 18s, 23 page views, currently viewing /product/widget-pro). Session #c5d1 (user ID 9102, 7m 41s, 15 page views, last active 3 minutes ago on /support). All sessions include referrer, device type, and geographic location. This is an illustrative summary—actual API returns full session objects with event timelines.
This example demonstrates the MCP's core list-and-filter capability. Requires a valid API key with read access to session data. Be aware that high-traffic sites may hit rate limits if querying frequently; the tool supports pagination but returns a fixed page size per call.
@Livesession find all sessions in the past hour where users visited the pricing page but didn't reach checkout
Found 18 sessions matching criteria: visited /pricing but no /checkout event in the past 60 minutes. Example: Session #d8k2 (user ID 4392, spent 4m 12s on /pricing, then exited to /features). Session #e1m7 (user ID 5581, viewed /pricing twice, bounced after 2m 38s). Session #f9p3 (user ID 6704, /pricing → /faq → exit). Each session includes full navigation path, timestamps, and device context. This illustrative output shows how filtering by page path reveals drop-off patterns.
Showcases the MCP's ability to filter sessions by URL patterns and time windows, useful for conversion funnel analysis. The tool does not modify session data—it's read-only. Complex multi-condition filters may require multiple API calls; consider pagination if your site has high session volume.
@Livesession pull session data for users who triggered a JavaScript error in the last 24 hours and summarize the most common error types
Retrieved 34 sessions with JavaScript errors in the past 24 hours. Most common error: 'TypeError: Cannot read property of undefined' (14 occurrences, primarily on /dashboard). Second: 'ReferenceError: gtag is not defined' (9 occurrences, /blog pages). Third: 'NetworkError: Failed to fetch' (6 occurrences, /api/data endpoint). Sessions include full error stack traces, browser versions (Chrome 120: 18 sessions, Firefox 121: 11 sessions, Safari 17: 5 sessions), and user actions leading to errors. This is an illustrative synthesis—actual data depends on your site's error tracking setup.
Demonstrates pairing the MCP's session retrieval with AI-driven pattern analysis. Useful for debugging production issues by correlating errors with user behavior. Requires that your Livesession account captures JavaScript errors (not all plans include this). The AI summarizes trends; you'll need to inspect individual sessions for root-cause details.
Use-case deep-dives
When Livesession MCP helps support teams debug faster
A 3-person support team handling 40-60 tickets daily needs to pull session replays when customers report UI bugs. The Livesession MCP lets them query sessions by user ID or timestamp directly from Switchy without opening the Livesession dashboard. This works when your support flow already lives in a shared workspace and you're filtering by known criteria—user email, date range, or session ID from a ticket. The single tool here is list-only; you still click through to Livesession's UI to watch the replay. If your team needs to search sessions by freeform text or event type, the MCP won't help—you'll need the full dashboard. Use this when your support reps already know which session they want and just need the link fast.
Livesession MCP for lightweight session audits
A solo product manager at a 12-person SaaS startup reviews 10-15 user sessions each Monday to spot onboarding friction. The Livesession MCP lets them pull the week's sessions filtered by new-user cohort without leaving their planning doc in Switchy. This is faster than logging into Livesession's dashboard when you're already writing notes in the same workspace. The trade-off: you can only filter by the criteria Livesession's API supports—date, user, duration. If you need to segment by feature usage or funnel step, the MCP won't surface those sessions; you'll still need the dashboard's advanced filters. This wins when your review cadence is weekly or slower and your filter needs are simple.
When Livesession MCP fits post-release monitoring
A 6-engineer team deploys to production twice a week and wants to spot broken flows in the first hour. The Livesession MCP can pull sessions from the last 60 minutes filtered by error state or specific user cohorts, surfacing issues before customers report them. This works when your deploy checklist already lives in Switchy and you want session data inline with your runbook. The limitation: the MCP lists sessions but doesn't expose event-level detail or heatmaps. If your smoke check needs to confirm click paths or form interactions, you'll still open the Livesession UI. Use this when your post-deploy ritual is lightweight and you're checking for obvious breakage, not deep funnel analysis.
Frequently asked
What does the Livesession MCP do in Switchy?
It pulls session replay data from your Livesession account into Switchy's AI workspace. The MCP lets you query active user sessions with filters like date range, user ID, or session properties. Your team can analyze session patterns, debug user issues, or generate support reports without leaving the conversation thread.
Do I need admin access to connect Livesession?
You need a Livesession API key, which typically requires account owner or admin permissions to generate. The key authenticates all requests, so whoever connects it in Switchy will grant the AI read access to session data visible under that key's scope. Check your Livesession account settings to confirm your role can create API keys.
Can the MCP watch sessions in real time or trigger recordings?
No. The MCP only lists sessions that Livesession has already recorded. It can't start new recordings, watch live sessions as they happen, or modify session metadata. If you need real-time monitoring, use Livesession's dashboard directly. The MCP is for pulling historical session data into AI workflows after the fact.
How is this different from exporting session data manually?
Manual exports require you to log into Livesession, apply filters, download CSVs, then paste results into a doc or Slack. The MCP lets your team ask natural-language questions like "show me sessions from users in France yesterday" and get structured answers instantly. It's faster for ad-hoc analysis and keeps session insights inside your team's AI context.
Should our support team or engineering team connect this?
Whoever debugs user issues most often. Support teams use it to pull session replays when investigating tickets. Engineering teams use it to correlate sessions with error logs or feature usage. One connection works for everyone in the workspace, so connect it under the role that owns your Livesession account and needs the broadest session visibility.