Scoring was not performed

• CRITICAL: Add explicit input schemas to jira_createIssue, jira_updateIssue, and jira_transitionIssue. At minimum, jira_transitionIssue needs issueKey and transitionId with type definitions and constraints.
• CRITICAL: Document output schemas for ALL 28 tools. Create a canonical response schema per tool showing field names, types, and which fields are suitable for chaining into downstream tools. Example: bitbucket_getProject returns {projectKey, projectName, public, links}.
• HIGH: Add enum constraints to all enum-like parameters. For bitbucket_getPullRequests.state, document: 'enum: ["OPEN", "DECLINED", "MERGED"]'. For bitbucket_getPullRequestDiff.diffType, document valid options.
• HIGH: For all parameters with numeric ranges (start, limit, line), document minimum and maximum explicitly. Example: 'limit (1-100, default 20)' instead of just 'Pagination limit'.
• HIGH: Add confirmation/dry-run support to write operations. Either add a 'dryRun' boolean parameter to create/update tools, or split write tools into a 'preview' variant that returns what would be created/changed without committing.
• MEDIUM: Refactor bitbucket_getPR_CommentsAndAction into two separate tools: bitbucket_getPullRequestComments and bitbucket_getPullRequestActions. Clarifies intent and makes LLM selection unambiguous.
• MEDIUM: Add dependency hints to tool descriptions. Example in bitbucket_updatePullRequest: 'To add reviewers without removing existing ones, first call bitbucket_getPullRequest to fetch current reviewers.'