Advanced security scanner for Model Context Protocol services with multi-tool orchestration
Scoring was not performed
Server purpose mismatch: pyproject.toml describes 'Advanced security scanner for Model Context Protocol services' but tools are time utilities. Tools do not match server domain.
Tool descriptions are generic and minimal (under 60 chars). 'Get current time in a specific timezones' lacks actionable context for LLM selection. Does not explain WHEN to use vs other tools, prerequisites, or return structure.
Parameter description for 'timezone' contains example values ('America/New_York', 'Europe/London') in natural language instead of declaring a constrained enum or pattern. LLMs may hallucinate invalid timezone strings.
No output schema documented. LLMs do not know what fields to expect from these tools (e.g., does get_current_time return {time: string}? {timestamp: number}? {datetime: string, unix_epoch: number}?).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 23 | <=2025-11-25 | v2 |
| 2026-03-09 | F | 48 | - | v1 |
No error handling guidance. If an invalid timezone is passed, what error does the tool return? How should the LLM recover?
Parameter 'time' in convert_time expects '24-hour format (HH:MM)' but this constraint should be a regex pattern or enum, not buried in description text. LLMs cannot parse informal descriptions as machine constraints.
No scanning tools visible despite server claiming to be a security scanner. Either tools are missing from this scan or the server definition is incomplete.