MCP server for interacting with SkySQL database services, including provisioning serverless databases, managing agents, and executing queries
Server has 4 tools with basic schemas and descriptions. Naming follows verb_noun convention (list_agents, launch_serverless_db, delete_db, ask_agent). Descriptions are present but brief (10-50 chars). Input schemas are visible with type definitions for all parameters. However, output schemas are not documented, parameter descriptions lack detail about constraints/formats, and error handling guidance is absent. No enum constraints despite tools accepting specific values (region, provider). Missing documentation of what each tool returns and when to use it vs alternatives.
Ask a question to a specific DB agent
Delete a DB instance from SkySQL
Launch a new Serverless DB instance in SkySQL
List all available SkySQL DB agents with their capabilities
Output schemas not documented. Tools return structured objects (ServerlessDBResponse, AgentInfo, LlamaResponse defined in code) but descriptions do not explain what fields are returned or their meaning. LLMs cannot plan downstream calls without knowing response structure.
Parameter descriptions lack constraint details. 'region' and 'provider' parameters have defaults but no enum values listed, no validation rules, and no guidance on valid options. LLMs will guess or hallucinate values.
No error handling guidance. Tools do not document what errors can occur, whether they are retryable, or what the LLM should do next. A failed delete_db call provides no recovery path.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 64 | <=2025-11-25 | v2 |
Destructive operations lack confirmation step. delete_db and launch_serverless_db modify state but have no dry-run or confirmation mechanism. Agents can accidentally destroy resources.
Tool descriptions too brief (10-50 chars). 'List all available SkySQL DB agents with their capabilities' (54 chars) lacks context on WHEN to call this vs ask_agent, what structure is returned, and whether results are paginated. Baseline for A+ tools is 50-200 chars with clear WHAT/WHEN/RETURN guidance.