MCP server for OSPREY framework - an agentic interface to scientific control systems with tools for setup inspection, configuration patching, and workspace management
Single tool 'setup_inspect' has a detailed description (287 chars, within 10-1024 range) explaining what it does and security considerations (masking secrets). However, the input schema is empty ({}), which is a critical violation. The tool lacks parameter descriptions, type definitions, and output schema documentation. No error handling guidance provided. The description is good but cannot compensate for missing schema structure.
Inspect OSPREY agent configuration. Returns a comprehensive JSON snapshot of the current agent configuration including config.yml, MCP servers, settings, hooks, rules, agents, skills, environment variables, and workspace structure. The config is reported UNEXPANDED --- ``${VAR}`` placeholders intact --- so a resolved secret never reaches the transcript. Environment variables and any literal value in the config or ``.mcp.json`` whose key contains KEY, TOKEN, SECRET or PASSWORD are masked on top of that.
Input schema is empty ({}). No parameters defined, typed, or described.
No output schema documented. LLMs cannot plan downstream calls or extract fields from the JSON snapshot returned.
No error handling guidance. If config parsing fails or secrets masking encounters an error, the LLM has no recovery path.
Tool name 'setup_inspect' does not start with a clear action verb. 'inspect_agent_config' or 'get_agent_configuration' would be clearer.
Description mentions 'config.yml', 'MCP servers', 'hooks', 'rules', 'agents', 'skills', 'environment variables', 'workspace structure' but does not document which fields are present in the returned JSON or their types.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |