feat: Add tool annotations for improved LLM tool understanding

[triepod-ai]

Summary

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

Changes

• Added readOnlyHint: true to qdrant-find (read-only search tool)
• Added destructiveHint: true to qdrant-store (data modification tool)
• Added title annotations for human-readable display

Why This Matters

• Annotations provide semantic metadata that helps LLMs understand tool behavior
• LLMs can make better decisions about when to use tools and in what order
• Enables safer tool execution by distinguishing read-only from destructive operations

Testing

• All 24 tests pass
• Ruff linter passes
• Live verification: Created server instance and confirmed tools/list returns annotations:

  Tool Name            Title                ReadOnly   Destructive
  ------------------------------------------------------------
  qdrant-find          Find in Qdrant       True       N/A       
  qdrant-store         Store in Qdrant      N/A        True      
  

Before/After

Before:

self.tool(
    find_foo,
    name="qdrant-find",
    description=self.tool_settings.tool_find_description,
)

After:

self.tool(
    find_foo,
    name="qdrant-find",
    description=self.tool_settings.tool_find_description,
    annotations=ToolAnnotations(
        title="Find in Qdrant",
        readOnlyHint=True,
    ),
)

🤖 Generated with Claude Code