Coassemble
Coassemble is a platform that allows users to create, manage, and deliver online training courses.
Verdict
Common use cases
- Check course completion rates for Q1
- List all users enrolled in compliance training
- Pull learner progress for performance reviews
- Audit which clients have active courses
- Find users who haven't completed onboarding
Integration
- Vendor
- Coassemble
- Category
- other
- Auth
- API_KEY
- Tools
- 4
- Composio slug
coassemble
Tools
- Get Clients
Tool to retrieve a paginated list of clients. use when you need to browse or process organisation records sequentially.
- Get Courses
Tool to retrieve all courses with optional filters and pagination. use when you need to list courses after setting identifier and client identifier, and control page size and number.
- Get Trackings
Tool to retrieve tracking objects with filters. use when you need to fetch learner progress data filtered by course id, date range, identifier, or completion status.
- Get Users
Tool to retrieve a paginated list of users. use when you need to browse or process user records optionally filtered by clientidentifier.
Setup
Setup guide
- 11. In Switchy, open your workspace settings and navigate to the MCP integrations panel. 2. Select 'Add MCP' and choose Coassemble from the list. 3. Log into your Coassemble account, go to Settings > API, and generate a new API key with read permissions for clients, courses, users, and tracking data. 4. Copy the key and paste it into the Switchy prompt, then click 'Connect'. 5. Switchy will verify the key and confirm the connection. 6. Open any Space, type '@Coassemble get courses' and send the message to test — you should see a list of your active courses returned in the chat. 7. If the response is empty or errors, double-check the API key has the correct scopes and hasn't expired in Coassemble.
What teammates see: by default, memories from Coassemble 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
Completion Report
@Coassemble get trackings for course ID 12345 filtered by completion status true, then summarize completion rate and list any users still pending.Open in a Space →
Active Courses Overview
@Coassemble get courses with no filters, then format the results as a table showing course name, ID, and client identifier.Open in a Space →
User Enrollment Check
@Coassemble get users filtered by client identifier ABC123, then list their names and email addresses in a numbered list.Open in a Space →
Client Audit
@Coassemble get clients with pagination set to 50 per page, then show the first page and tell me how many total clients exist.Open in a Space →
Incomplete Training Alert
@Coassemble get trackings for course ID 67890 where completion status is false and start date is after January 1 2024, then list the user IDs and how long since they last accessed the course.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Coassemble show me all active courses for our main training client, including enrollment counts
Based on the client identifier you've set, here are 8 active courses: "Onboarding Essentials" (47 enrolled), "Product Knowledge Q2" (23 enrolled), "Compliance Refresher" (89 enrolled), "Leadership Fundamentals" (12 enrolled), "Sales Techniques" (34 enrolled), "Customer Service Excellence" (56 enrolled), "Data Privacy Training" (102 enrolled), and "Time Management" (19 enrolled). All courses are currently published and accepting new learners.
This example demonstrates the Get Courses tool with client filtering. You'll need to know your client identifier beforehand (fetch it via Get Clients first). The response reflects a snapshot; enrollment numbers change as learners join or complete courses. Pagination applies if you have more than the default page size.
@Coassemble find all users who haven't completed the compliance course in the last 90 days
Checking tracking records for your compliance course over the past 90 days: 14 users have incomplete progress. Notable entries include Sarah Chen (45% complete, last activity 12 days ago), Marcus Williams (0% complete, enrolled 78 days ago), and Priya Patel (88% complete, last activity 3 days ago). The remaining 11 users range from 10% to 65% completion. All are still within the enrollment window.
This example uses Get Trackings with date range and completion filters. You must provide the course ID upfront. The tool returns raw tracking objects—the AI synthesizes completion percentages and last-activity timestamps into readable summaries. Be aware that "incomplete" includes users who just enrolled yesterday.
@Coassemble compare completion rates across all our Q1 courses and identify which topics need better engagement
Analyzing completion data across your 6 Q1 courses: "Product Launch Training" leads at 82% completion (avg. 4.2 days to finish), followed by "Security Basics" at 76% (3.8 days). However, "Advanced Analytics" shows only 34% completion with an average 18-day duration for those who finish—suggesting content complexity or length issues. "Team Collaboration" sits at 41%, with many users stalling around the midpoint. Consider breaking these into shorter modules or adding checkpoints.
This synthesis example chains Get Courses and Get Trackings, then applies AI reasoning to spot patterns. The analysis quality depends on having sufficient tracking history. Results reflect correlation, not causation—low completion might stem from course design, audience fit, or external factors the MCP can't see.
Use-case deep-dives
When Coassemble MCP makes quarterly L&D reporting faster
A 12-person ops team runs quarterly compliance training through Coassemble and needs to pull completion data into a shared Switchy workspace for board review. The Coassemble MCP is the right call here: the Get Trackings tool filters by completion status and date range, so you can pull Q3 completions in one query and drop them into a shared doc without exporting CSVs. The four-tool scope is narrow enough that the AI won't hallucinate extra capabilities, and API key auth means one admin can connect it for the whole team. If you're auditing more than 500 learners per quarter, expect to paginate manually—Coassemble's API doesn't batch well at scale. For teams running monthly compliance checks with under 100 users, this MCP saves 20 minutes of export-paste work every cycle.
Coassemble MCP works for multi-tenant training ops
A 6-person customer success team manages onboarding courses for 40 client accounts in Coassemble and needs to assign new hires to the right course catalog each week. The Coassemble MCP handles this scenario cleanly: Get Clients lists all orgs, Get Courses filters by client identifier, and Get Users pulls the roster for assignment checks. The workflow fits in a single Switchy thread—no need to context-switch between the LMS admin panel and Slack. The limitation is write access: this MCP is read-only, so you can't auto-enroll users or create courses from the AI workspace. If your onboarding process is mostly lookup and verification (not bulk enrollment), the MCP cuts 10 minutes per client setup. For teams doing weekly bulk enrollments, you'll still need the Coassemble UI.
When Coassemble MCP isn't the right fit for just-in-time learning
A 20-person sales org uses Coassemble to host product training videos and wants reps to ask an AI for course links during customer calls. The Coassemble MCP is borderline here: Get Courses can retrieve the catalog, but the tool doesn't return video URLs or lesson content—just metadata like course titles and IDs. If your use case is 'find the right course and open it in Coassemble,' the MCP works. If reps need the AI to surface specific lesson transcripts or embed training clips in Switchy, the MCP can't deliver that. The four-tool scope is too narrow for content retrieval. For teams where the training catalog changes weekly and reps just need quick course discovery, the MCP saves 5 minutes per lookup. For deeper content access, you're better off with a custom integration or Coassemble's native search.
Frequently asked
What does the Coassemble MCP do in Switchy?
It pulls learner data, course lists, and client records from your Coassemble LMS into Switchy's AI workspace. Your team can query enrolment progress, filter by completion status or date range, and surface user records without logging into Coassemble's dashboard. Useful when you need to report on training outcomes or audit course assignments across multiple clients.
Do I need admin access to connect Coassemble?
You need an API key from Coassemble, which typically requires account-owner or admin permissions to generate. Once you paste the key into Switchy, the MCP authenticates automatically. If your Coassemble plan doesn't expose API keys, you won't be able to connect this integration—check with your account manager first.
Can the Coassemble MCP create or edit courses?
No. The four tools are read-only: they retrieve clients, courses, tracking records, and users. You can't enrol learners, update course content, or change completion statuses through Switchy. For write operations, use Coassemble's web interface or a dedicated automation platform that supports their full API.
How is this different from exporting CSV reports from Coassemble?
Switchy's MCP lets you query live data conversationally—ask for incomplete courses in the last 30 days, filter by client identifier, or cross-reference user lists without downloading spreadsheets. CSV exports are static snapshots; the MCP pulls fresh records on demand and integrates with other tools in your workspace.
Who on my team should connect the Coassemble MCP?
Whoever holds the API key and needs to report on learner progress—usually an L&D manager, training coordinator, or ops lead. Once connected, anyone in your Switchy workspace can query the data. The connection doesn't consume per-seat licences in Coassemble; it counts as a single API client.