Skip to main content
Glama
524,045 tools. Updated 2026-09-06 14:24

"Qase" matching MCP tools:

  • Publish announcements to an Oase feed with markdown. Optional title, defaults to last joined Oase, returns post ID for comments and reactions.
    MIT
  • Call any Qase REST endpoint directly when no dedicated tool exists. Pass HTTP method, /v1/ path, and optional body or query to get raw API responses.
    MIT
  • Search by action keywords to find and activate hidden Qase tools, including deletes, reviews, and custom fields. Makes these capabilities callable when they're not visible by default.
    MIT
  • Get all project context in one call—details, suites, milestones, environments, custom fields, users. Use as the first step to bootstrap queries; inspect coverage and set full to retrieve complete collections.
    MIT
  • File a defect from a test failure with the failure context embedded as the actual result. Requires title, actual_result, and severity to create the defect.
    MIT
  • Search Qase entities using QQL to filter, aggregate, and fetch many records at once. Ideal for questions like 'failed results this week' or 'open blockers across projects'.
    MIT

Matching MCP Servers

Matching MCP Connectors

  • Official MCP server for Qase — manage test cases, runs, suites, defects via AI tools.

  • MarkupBase turns AI-generated Markdown and HTML into durable, versioned artifacts that people can review and discuss. Its MCP server lets agents publish new versions, preserve contextual comments, include hosted images, and respond to feedback through secure account-linked identities, creating a clear human review boundary without requiring real-time editing.

  • Update a post on an Oase feed with new markdown body content. Provide the post_id and new body; optionally adjust the title or oase_id.
    MIT
  • Create or update a tracked testing defect in Qase. New defects require title, actual_result, and severity; include an existing id to update or resolve it.
    MIT
  • Create a new test run or update an existing one by providing an ID. Use it to open a run before recording test results, with optional scoping to milestone, environment, plan, or cases.
    MIT
  • Retrieve decrypted recent posts from an Oase feed, ordered oldest first, with post IDs and attachments for further actions like commenting or reacting.
    MIT
  • Upload a file (base64 or local path) and get a hash to attach it to Qase cases, results, or defects. Use the returned hash in attachment fields instead of re-uploading.
    MIT
  • Create or update a single Qase test case: include an ID to update, omit it to create; supports steps, attachments, enum fields. For multiple cases, use bulk create.
    MIT
  • Starts persistent Promise login for Claude by opening browser and launching a local server, returns URL as fallback.
    MIT
  • Fetch a single Qase entity (case, suite, run, result, plan, defect, etc.) by its ID and type. Use when you already know the exact ID and need one record; for multiple or filtered records, use qql_search instead.
    MIT
  • Edit your own previously sent chat message by replacing its text with new markdown, keeping attachments. Provide the message ID from send or read operations.
    MIT
  • Delete a post from an oase feed by providing its post_id. Use for your own posts, or with admin/owner access to remove others' posts.
    MIT
  • Retrieve recent chat messages from an encrypted oase, ordered oldest first. Includes message IDs, reply threading, and reaction tags to guide responses.
    MIT
  • Complete the Oase login by waiting for the token from the browser redirect, exchanging it for a persistent identity, and shutting down the callback server.
    MIT
  • Delete a chat message by providing its message_id. Optionally specify the oase_id to target a specific chat; defaults to the last joined oase. The message is soft-deleted, showing as removed.
    MIT
  • React to a message with an emoji by providing its ID. One reaction per message per participant.
    MIT
  • Report a finished CI run in one call: create the run, record all test results, and complete it, avoiding half-open runs.
    MIT