Seed AI catalog with top MCP servers for use in DAP agents and flows
<!--IssueSummary start--> <details> <summary> Everyone can contribute. [Help move this issue forward](https://handbook.gitlab.com/handbook/marketing/developer-relations/contributor-success/community-contributors-workflows/#contributor-links) while earning points, leveling up and collecting rewards. </summary> - [Close this issue](https://contributors.gitlab.com/manage-issue?action=close&projectId=278964&issueIid=591969) </details> <!--IssueSummary end--> ## Summary Customers have existing tool investments they won't abandon. Seeding the catalog with the tools they already use lowers adoption friction, demonstrates "better together" value, and gives users something meaningful to work with on day one. We want to seed the catalog with a curated set of high-quality, widely-used MCP servers that GitLab users can immediately add to their agents and flows. ## Scope The table below covers MCP server candidates grouped by type and ranked by priority within each group. Engineering should aim to include at least one server per group, starting with the highest priority candidate in each. This issue covers adding these servers to the catalog. Governance controls, vetting workflows, and contributor attribution are tracked separately. | Group | Priority | MCP server | Example use cases | Notes | |---|---|---|---|---| | **Issue tracking & team collaboration** | 1 | Jira/Confluence | 1. Sync Jira tickets to GitLab issues and MRs (e.g. close Jira ticket when MR merges) 2. Pull sprint/acceptance criteria context into code generation prompts | Near-universal in enterprise; official GA server from Atlassian with governed access | | | 2 | Slack | 1. Post CI/CD failure summaries or deployment notifications to the right channel 2. Pull thread context into incident postmortems or MR descriptions | High-value for escalation and notification flows | | | 3 | Figma | 1. Generate frontend code scaffolding from Figma component specs 2. Link design token changes to MR reviews | Narrower audience but strong for full-stack teams | | | 4 | GitHub | 1. Manage issues, PRs, and repo context for teams in hybrid GitLab + GitHub environments 2. Mirror or migrate repository workflows to GitLab | Most strategically sensitive pick — customers in multi-VCS environments need this; recommend leadership alignment before shipping | | | 5 | Salesforce | 1. Link Salesforce opportunities and cases to GitLab issues and MRs for sales-to-engineering traceability 2. Pull customer context into incident or feature prioritization flows | Near-universal in enterprise; supports "better together" for teams connecting customer activity to engineering work | | | 6 | Google Calendar | 1. Surface team availability and OOO context for agents making assignment, escalation, or scheduling decisions across the SDLC 2. Gate or inform deployment and release timing based on team coverage and calendar state | Validated by internal Create team request for intelligent reviewer assignment (#592752); broadly applicable anywhere agents need people-availability context | | **Cloud & infrastructure** | 1 | AWS | 1. Automate deployments and inspect CloudWatch logs post-deploy 2. Generate and validate infrastructure-as-code from natural language prompts | Dominant cloud provider in GitLab's enterprise base | | | 2 | Terraform | 1. Generate and review Terraform plans as part of CI/CD pipelines 2. Detect and explain infrastructure drift | Strong overlap with existing GitLab CI/CD workflows; IaC is table stakes | | | 3 | Kubernetes | 1. Inspect pod/cluster health as part of deployment verification flows 2. Trigger rollbacks or scaling actions in response to CI/CD events | Runtime deployment target for most containerized workloads | | | 4 | Azure DevOps | 1. Sync work items and pipelines for teams in hybrid GitLab + Azure environments 2. Migrate pipeline definitions to GitLab CI/CD | Relevant for Microsoft-shop enterprises; partially competitive with GitLab — align on positioning | | **Security & code quality** | 1 | Snyk | 1. Surface dependency vulnerabilities in MR context with fix suggestions 2. Block or flag pipelines based on Snyk severity thresholds | Developer-first security brand; customers have it deployed and won't rip it out — validates "better together" positioning | | | 2 | Sentry | 1. Bring live error context into agents to generate fix suggestions or MRs automatically 2. Link Sentry alerts to GitLab issues for triage | Strong remediation story; bridges monitoring to code | | | 3 | JFrog | 1. Scan artifacts for vulnerabilities before promotion in release pipelines 2. Enforce supply chain policies as part of CI/CD gates | High relevance for enterprise compliance; narrower audience than Snyk | | **Observability & incident management** | 1 | Datadog | 1. Query APM and log data post-deploy to detect regressions automatically 2. Trigger GitLab rollback flows based on Datadog anomaly alerts | Dominant observability platform in enterprise; strongest brand and widest deployment | | | 2 | Grafana | 1. Embed dashboard context into incident postmortems generated by agents 2. Query metrics to validate deployment health in flows | Strong in cloud-native and open source shops; complements Prometheus setups | | | 3 | PagerDuty | 1. Auto-create PagerDuty incidents from CI/CD failures or deployment events 2. Pull on-call context into GitLab incident issues | Natural fit for ops teams managing deployment-related incidents | | **Data & analytics** | 1 | Snowflake | 1. Query Snowflake as part of data pipeline CI/CD validation flows 2. Generate and test SQL transformations from natural language prompts | Only data platform candidate in this group; relevant for data engineering teams using GitLab for pipeline CI/CD |
issue