AirOps MCP
AirOps MCP lets agents access authorized AirOps brand context, knowledge bases, content workflows, and tracked Brand Kit editing capabilities.
Verdict
Common use cases
- Trigger lead enrichment workflows from Slack threads
- Check pipeline run status during standups
- Query processed datasets for customer insights
- Kick off content generation batches on demand
- Pull workflow results into project briefs
Integration
- Vendor
- AirOps MCP
- Category
- other
- Auth
- OAUTH2
- Composio slug
airops_mcp
Tools
Setup
Setup guide
- 11. Open your Switchy workspace settings and navigate to the Integrations page. 2. Find AirOps MCP in the list and click Connect. 3. You'll be redirected to AirOps to sign in and authorize Switchy — grant access to workflows, datasets, and app execution. 4. After authorization, you'll return to Switchy with the connection active. 5. Open any Project and type '@AirOps' to confirm the integration appears in the mention menu. 6. Test the connection by asking '@AirOps list my workflows' or '@AirOps show recent pipeline runs' — you should see live data from your AirOps workspace. 7. If the connection fails, revisit Integrations to check OAuth status and re-authorize if needed.
What teammates see: by default, memories from AirOps MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
List Active Workflows
@AirOps list all active workflows in my workspace and show their last run status
Trigger Enrichment Pipeline
@AirOps trigger the lead enrichment workflow for the contacts uploaded today
Check Pipeline Status
@AirOps check the status of the content generation pipeline I started this morning
Query Dataset Results
@AirOps retrieve the results from the last customer segmentation workflow and summarize key findings
Review Recent Runs
@AirOps show me the last five workflow runs and flag any that failed or timed out
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@AirOps search our knowledge base for recent customer feedback about the new onboarding flow and summarize the top three themes
Based on 47 knowledge base entries from the past two weeks, three themes emerge: (1) Users appreciate the reduced step count but find the progress indicator unclear (mentioned in 18 entries). (2) The email verification step causes confusion when users don't receive the code immediately (12 entries). (3) Mobile users report layout issues on screens smaller than 375px (8 entries). The remaining feedback covers minor UI polish and feature requests. This search covered your 'Product Feedback' and 'Support Tickets' collections.
This example assumes AirOps indexes internal documentation or support data. The MCP requires OAuth scope to access your workspace's collections. Results depend entirely on what you've connected to AirOps — if your knowledge base is empty or siloed, the search returns nothing useful.
@AirOps create a new workflow that triggers when a Slack message contains 'urgent bug' and posts a summary to our #engineering channel
Workflow created: 'Urgent Bug Notifier'. Trigger: Slack message contains 'urgent bug' (case-insensitive). Action: Extract message context, generate a two-sentence summary using GPT-4, post to #engineering with a link to the original thread. Workflow ID: wf_8x3k9m2. Status: Active. You can edit this workflow in the AirOps dashboard under Automations > Workflows. First run will occur on the next matching message.
This demonstrates AirOps' automation capabilities. The MCP must have write access to both your Slack workspace and AirOps workflows. Side-effect risk: the workflow runs immediately after creation, so test in a sandbox channel first. Rate limits apply if your Slack generates high message volume.
@AirOps analyze the last 30 days of support tickets and identify which product areas generate the most repeat questions
Analysis complete across 312 support tickets (Oct 15 – Nov 14). Top repeat question areas: (1) Billing & invoices (89 tickets, 28.5%) — primarily about proration logic and receipt downloads. (2) API authentication (54 tickets, 17.3%) — token expiration and scope errors dominate. (3) Export functionality (41 tickets, 13.1%) — CSV formatting and timezone handling. The remaining 128 tickets span 14 other categories. This analysis assumes tickets are tagged; untagged tickets were excluded.
This example highlights AirOps' ability to synthesize patterns from structured data. Accuracy depends on consistent ticket tagging in your source system (Zendesk, Intercom, etc.). The MCP reads historical data but doesn't modify tickets. If your tagging is inconsistent, results will be skewed or incomplete.
Use-case deep-dives
When OAuth-gated AI workflows need centralized knowledge access
A 6-person support team routes Intercom tickets through Switchy, pulling context from internal docs and past resolutions. AirOps MCP fits if your knowledge base already lives in AirOps and your team authenticates via OAuth. The integration lets agents query that corpus without copying data into Switchy's workspace. The trade-off: if your docs are in Notion or Google Drive, you're adding a layer instead of removing one. This works when AirOps is already your source of truth and you need secure, scoped access across multiple agents. If you're starting from scratch, evaluate whether the OAuth overhead justifies the centralization versus lighter-weight MCP options.
Centralized prompt management for outbound workflows
A 4-person sales team runs personalized outreach sequences where each rep uses Switchy agents to draft emails, research prospects, and log activity. AirOps MCP makes sense if you're managing dozens of prompt templates across reps and need version control or A/B testing at the prompt level. The OAuth requirement means each rep authenticates once, then agents inherit scoped access to shared prompt libraries. The boundary: if your team is under 5 people and prompts change weekly, the setup cost outweighs the benefit. This pays off when you have 10+ reps, stable prompt patterns, and a need to audit which agent used which template. Smaller teams should start with Switchy's native prompt sharing.
When AirOps already powers your internal tools
A 3-person product team uses AirOps to automate user feedback tagging and feature request routing. Adding the MCP to Switchy lets agents trigger those workflows mid-conversation without switching contexts. This is the right call if AirOps is already embedded in your stack and you're paying for it. The OAuth layer ensures agents can't overwrite production workflows accidentally. The threshold: if you're not using AirOps elsewhere, this integration adds vendor lock-in for workflows you could build with lighter MCPs (Linear, Slack, Airtable). Use this when AirOps is already your automation backbone and you need Switchy agents to read or write to those pipelines. Otherwise, start with single-purpose MCPs.
Frequently asked
What does the AirOps MCP do in Switchy?
The AirOps MCP connects your Switchy workspace to AirOps' workflow automation platform. Your team can query AirOps data, trigger workflows, and pull execution results directly into AI conversations without switching contexts. This works best for teams already running AirOps automations who want their AI agents to interact with those processes.
Does AirOps MCP require OAuth or can I use an API key?
AirOps MCP uses OAuth2 authentication. You'll authorize Switchy through AirOps' standard OAuth flow, which means you need an AirOps account with appropriate workspace permissions. The connection lives at the workspace level in Switchy, so one team member can set it up and everyone inherits access based on their Switchy role.
Can the AirOps MCP create new workflows or just read existing ones?
This depends on the specific tools AirOps exposes through their MCP implementation. Most workflow platforms limit MCPs to read-only access and execution triggers for safety reasons. If you need to build new workflows, you'll likely still do that in the AirOps UI and then reference them by ID through Switchy.
Why use AirOps through Switchy instead of their API directly?
Switchy's MCP layer handles authentication, rate limiting, and error translation automatically. Your AI agents can call AirOps workflows using natural language instead of constructing API requests with the correct endpoint structure and parameters. For teams without dedicated developers, this removes the integration burden entirely.
Who on my team should connect the AirOps MCP?
Whoever owns your AirOps workspace should set up the connection. They need OAuth authorization rights in AirOps and workspace admin permissions in Switchy. Once connected, you can control which Switchy team members can invoke AirOps tools through standard role-based access controls. The connection itself doesn't consume additional AirOps seats.