Model Context Protocol server for the Bitrise CI/CD platform, providing tools to manage apps, builds, workspaces, artifacts, webhooks, pipelines, caching, and release management
The Bitrise MCP server provides 5 well-structured tools with generally solid naming conventions (all verb-noun style) and comprehensive input schemas. Tool descriptions are present and informative, ranging from ~90-210 characters. Parameter schemas are detailed with type information, enums, and descriptions. However, there are critical gaps: (1) Output schemas are not documented for any tool, the rubric requires 'Document the output schema' for tools to receive full credit; (2) Error handling guidance is absent, no recovery hints, categorization, or actionable error messages; (3) Some parameter descriptions are generic or lack constraint details (e.g., 'id' in create_connected_app just says 'uuidV4' without saying it's auto-generated if omitted); (4) The 'delete_app' description lacks severity warning and doesn't mention dry-run/confirmation patterns for an irreversible operation. The server demonstrates good fundamentals but falls short of production-grade tool design on critical dimensions.
Add a new Release Management connected app to Bitrise. Adding an internal tester grants the App Tester role on the app; removing them later does not revoke it.
Delete an app from Bitrise. When deleting apps belonging to multiple workspaces always confirm that which workspaces' apps the user wants to delete.
Gives back a Release Management connected app for the authenticated account.
List Release Management connected apps available for the authenticated account within a workspace.
Updates a connected app.
No output schemas documented. The rubric requires 'Document the output schema' (DIMENSION 1.D) so LLMs know what fields to expect and can plan downstream calls. Currently impossible to verify what create_connected_app, update_connected_app, or delete_app return.
delete_app lacks destructive operation warnings and confirmation/dry-run support. The rubric states: 'Irreversible operations (delete, send, publish) should support a dry-run or confirmation step' (DIMENSION 1.E). Deleting an app is catastrophic and should require explicit confirmation or offer a dry-run preview.
No error handling guidance. The rubric requires 'Error responses must tell the LLM what to do next' (DIMENSION 1.E). Tools do not document what errors can occur, how to categorize them (retryable vs user-fixable vs fatal), or what recovery steps to suggest.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | B | 72 | 2026-07-28+ | v2 |
| 2026-03-09 | D | 57 | - | v1 |
Parameter 'id' in create_connected_app lacks clarity on auto-generation. Description says 'An uuidV4 identifier for your new connected app. If it is not given, one will be generated.' This is helpful but should explicitly state 'optional; server generates if omitted' to guide LLM behavior.
delete_app description is vague and lacks pre-requisite warnings. Current: 'Delete an app from Bitrise. When deleting apps belonging to multiple workspaces always confirm that which workspaces' apps the user wants to delete.' This does not explain consequences, irreversibility, or multi-workspace implications clearly enough for safe LLM usage.
create_connected_app parameter 'tester_groups' and 'code_push_deployments' are complex nested arrays but lack examples or constraint documentation. The rubric notes descriptions should be 'as if prompt-engineering', these params would benefit from explicit guidance on structure and validation.
list_connected_apps response is not documented. Does it return total_count? next_cursor? The rubric requires 'Tools returning lists should accept page/offset and limit parameters and return a total count or next_cursor' (DIMENSION 1.D). Without documented pagination metadata in responses, agents cannot reliably iterate large result sets.