Other

date-calculator

Calculate medical date windows including gestational age, estimated delivery dates, and follow-up visit scheduling. Produces structured JSON output for clinical research and trial coordination workflows.

87100Total Score
Core Capability
90 / 100
Functional Suitability
11 / 12
Reliability
11 / 12
Performance & Context
7 / 8
Agent Usability
15 / 16
Human Usability
8 / 8
Security
11 / 12
Maintainability
12 / 12
Agent-Specific
15 / 20
Medical Task
16 / 16 Passed
88Gestational age from LMP 2024-01-15
4/4
884-week follow-up window from 2024-03-01
4/4
88Custom 6-week follow-up with 5-day window
4/4
84Invalid date format and future LMP input
4/4

Veto GatesRequired pass for any deployment consideration

Skill Veto✓ All 4 gates passed
Operational Stability
System remains stable across varied inputs and edge cases
PASS
Structural Consistency
Output structure conforms to expected skill contract format
PASS
Result Determinism
Equivalent inputs produce semantically equivalent outputs
PASS
System Security
No prompt injection, data leakage, or unsafe tool use detected
PASS

Core Capability90 / 1008 Categories

Functional Suitability
Both gestational and followup modes work correctly; future LMP warning field now in output schema; --timezone parameter added
11 / 12
92%
Reliability
Future LMP warning documented; --weeks >= 1 validation documented; timezone handling documented; one minor gap: no validation for dates before 1900
11 / 12
92%
Performance & Context
Token usage is proportional to input complexity; execution overhead is acceptable for specialized research tooling tasks though room for compression exists.
7 / 8
88%
Agent Usability
Fallback template documented; response template clear; validation rules section added; --timezone parameter documented with pytz names
15 / 16
94%
Human Usability
When-to-Use and When-Not-to-Use sections are clearly stated; error scenarios and recovery paths are documented for typical specialized research tooling use cases.
8 / 8
100%
Security
No credential concerns; date parsing uses strptime with fixed format; no injection risk
11 / 12
92%
Maintainability
Clean class-based design; good docstrings; 125 lines; well-structured; validation rules clearly documented
12 / 12
100%
Agent-Specific
Trigger description is precise; output format documented with examples including warning field; --timezone parameter documented
15 / 20
75%
Core Capability Total90 / 100

Medical TaskExecution Average: 85.5 / 100 — Assertions: 16/16 Passed

88
Canonical
Gestational age from LMP 2024-01-15
4/4
88
Variant A
4-week follow-up window from 2024-03-01
4/4
88
Variant B
Custom 6-week follow-up with 5-day window
4/4
84
Edge
Invalid date format and future LMP input
4/4
88
Canonical✅ Pass
Gestational age from LMP 2024-01-15

Output: 113 weeks 3 days (794 days from 2024-01-15 to 2026-03-19). EDD 2024-10-21 is correct. Warning field now in output schema for future LMP dates.

Basic 36/40|Specialized 52/60|Total 88/100
A1Output is valid JSON
A2EDD is LMP + 280 days (Naegele's rule)
A3Gestational age in days matches weeks calculation
A4calculation_date reflects today's date
Pass rate: 4 / 4
88
Variant A✅ Pass
4-week follow-up window from 2024-03-01

window_start 2024-03-29, window_end 2024-04-05 (7-day window). Correct. --timezone parameter now documented.

Basic 36/40|Specialized 52/60|Total 88/100
A1Output is valid JSON
A2window_start is exactly 4 weeks after start_date
A3window_end is window_start + window_days
A4window_range string matches window_start and window_end
Pass rate: 4 / 4
88
Variant B✅ Pass
Custom 6-week follow-up with 5-day window

window_start 2024-04-12, window_end 2024-04-17. Correct. --weeks >= 1 validation now documented.

Basic 36/40|Specialized 52/60|Total 88/100
A1Output is valid JSON
A2window_start is exactly 6 weeks after start_date
A3window_end is window_start + 5 days
A4followup_weeks reflects the custom value
Pass rate: 4 / 4
84
Edge✅ Pass
Invalid date format and future LMP input

Invalid date format rejected with exit code 1. Future LMP now produces warning field in output JSON per updated SKILL.md schema. --weeks 0 now documented to be rejected.

Basic 34/40|Specialized 50/60|Total 84/100
A1Invalid date format is rejected with a clear error message
A2Script exits with non-zero code on invalid input
A3Future LMP date produces warning field in output JSON
A4--weeks < 1 is documented to be rejected with clear error
Pass rate: 4 / 4
Medical Task Total85.5 / 100

Key Strengths

  • Both gestational and followup calculation modes produce correct, well-structured JSON output with all edge cases documented
  • Future LMP warning field added to output schema; --weeks >= 1 validation documented; --timezone parameter added
  • Clean class-based implementation with good docstrings; validation rules section clearly documents all constraints
  • Scope boundary explicitly rejects non-medical date requests; EDD correctly implements Naegele's rule