core-agent exposes only 2 tools with minimal structure. Both tools have descriptions, but they are vague and do not follow agentic tool patterns. 'return_result' and 'report_done' appear to be lifecycle/control tools rather than domain-specific capabilities. Input schemas are present but lack proper type definitions and parameter descriptions. The tools do not follow verb_noun naming conventions (e.g., 'return_result' is ambiguous, does it return to caller, to a database, to a log?). No output schemas are documented. Error handling is absent. The server appears designed for agent-to-agent delegation rather than exposing reusable, composable tools for LLM agents.

• Rename tools to follow verb_noun pattern: 'return_result' → 'submit_result' or 'complete_task'; 'report_done' → 'mark_complete' or 'finish_task'. Names should convey action, not control flow.
• Expand descriptions to 100 - 200 characters explaining WHAT each tool does, WHEN to call it, and what it returns. Example: 'submit_result: Send findings back to the delegating agent and terminate this task. Use when analysis is complete and you have evidence to report.'
• Add explicit JSON Schema type definitions to all parameters. Change 'result' from bare description to: {"type": "string", "description": "Your findings: the answer, analysis, or proposed change, with evidence behind it. Minimum 50 characters."}
• Document output schemas for both tools. What does return_result return to the caller? What does report_done return? Include field names, types, and examples.
• Add error handling guidance: 'If submission fails, retry up to 3 times. If all retries fail, log the error and notify the delegating agent via [mechanism]. Do not silently drop results.'