Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. listSessions has 3 parameters with descriptions; createSession has 7 parameters, but several lack detail on constraints, formats, or dependencies. Neither tool documents return types or pagination behavior. Descriptions are present but generic (avg ~50 chars), below the 194-char baseline for production tools. No error handling guidance, no idempotency hints, no security annotations. Tool names follow verb_noun convention (positive), but parameter naming is inconsistent (agentId vs agent_id). Output schemas are not visible in source.

• Document return types for both tools. listSessions should specify: {sessions: [{id, agentId, title, createdAt, status}], total, nextPageToken}. createSession should specify: {sessionId, status, executionUrl}.
• Add min/max bounds to numeric parameters: limit (1-100, default 50). Add length/format constraints to string parameters: prompt (1-10000 chars), title (1-256 chars).
• Expand parameter descriptions to 50-100 chars each. E.g., 'vaultIds: Optional array of credential vault IDs to inject into the agent environment. Each ID must be a valid UUID.'
• Add error handling guidance: 'If agentId is invalid, returns 404 with message "Agent not found. Call list_agents to see available agents." Retry is safe.'
• Standardize parameter naming to snake_case (agent_id, page_token, environment_id, vault_ids) for consistency with MCP conventions and to reduce LLM confusion.