feat: Add tool annotations for improved LLM tool understanding

[bryankthompson]

Summary

Adds MCP tool annotations (readOnlyHint, destructiveHint) to all 51 tools to help LLMs better understand tool behavior and make safer decisions about tool execution.

Changes

• Added readOnlyHint: true to 26 read-only tools (queries, fetches, searches)
• Added destructiveHint: true to 25 tools that create/update/delete data
• Added title annotations for human-readable display

Tool Breakdown

Category	Read-Only	Destructive	Total
Confluence	8	5	13
Jira	18	20	38
Total	26	25	51

Why This Matters

• Annotations provide semantic metadata that helps LLMs understand tool behavior
• readOnlyHint tells the LLM a tool is safe to call without side effects
• destructiveHint signals the LLM should be more careful before executing
• title provides a human-readable name for better context

Testing

• Server builds successfully (npm run build)
• TypeScript compilation passes with no errors
• All 51 tools have annotations added

Before/After

Before:

{
  name: "jira_get_issue",
  description: "Get detailed information about a single Jira issue...",
  inputSchema: { ... }
  // No annotations
}

After:

{
  name: "jira_get_issue",
  description: "Get detailed information about a single Jira issue...",
  inputSchema: { ... },
  annotations: {
    title: "Get Issue",
    readOnlyHint: true
  }
}

🤖 Generated with Claude Code