Static and dynamic security scanner for Model Context Protocol servers
mcpscan is a security scanner for MCP servers, not an MCP server itself. The 12 'tools' listed are actually test fixtures (hostile/malformed MCP servers used to validate the scanner). They are not real tool definitions but rather targets for security analysis. Most fixtures have minimal or no descriptions, incomplete schemas, and deliberately violate MCP patterns to test scanner detection. This is not a production MCP server and should not be scored as one. The fixtures serve a testing purpose, not an agent-facing interface.
Undeclared, but callable.
Returns the record matching an id.
Returns counts for the project.
Returns the contents of a file inside the project directory.
Harmless.
Not an MCP server, test fixtures only. The 12 'tools' are deliberately malformed MCP servers used to validate the scanner's detection capabilities, not production tool definitions.
7 of 12 tools have no description or trivial descriptions (<20 chars). Violates pattern:tool-description.
Most tools have incomplete input schemas: 'type:object' with no properties or required fields. Schemas lack parameter descriptions and type constraints.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 34 | 2025-06-18+ | v2 |
Searches the project for a string.
Reports which account the server is configured to use.
Tool 'tool_1' violates naming convention. Generic names like 'tool_1' provide no semantic meaning to LLMs. Should start with action verb (get_, create_, search_, etc.).
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract data without knowing return structure. Violates pattern:tool and pattern:response-shaper.