explain
Deep explanation of complex code, files, or concepts. Routes to expert agents, uses structural search, generates mermaid diagrams. Triggers on: explain, deep dive, how does X work, architecture, data flow.
84
Quality
84%
Does it follow best practices?
Impact
75%
2.14xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Evaluation results
62%
39%Authentication Module Documentation
Structured markdown output with Mermaid diagrams
Overview section
100%
100%
Architecture section
37%
100%
Mermaid diagram present
0%
100%
Appropriate diagram type
0%
100%
How It Works section
62%
0%
Key Concepts section
0%
0%
Dependencies section
0%
90%
Design Decisions section
50%
100%
Related Code section
0%
0%
File:line references
40%
30%
See Also section
0%
100%
Persistence offer
0%
0%
Without context: $0.3119 · 1m 50s · 11 turns · 18 in / 6,947 out tokens
With context: $0.8764 · 3m 25s · 27 turns · 280 in / 6,661 out tokens
92%
51%Query Pipeline Function Analysis
Symbol detection, expert routing, and CLI tool usage
Symbol search documented
58%
100%
TypeScript expert routing
90%
100%
Fallback behavior documented
87%
100%
Related files search
62%
100%
Overview section
100%
100%
Architecture section
0%
100%
Mermaid diagram included
0%
100%
How It Works section
0%
100%
Dependencies section
0%
0%
Design Decisions section
100%
100%
File:line references
0%
100%
See Also section
0%
100%
Without context: $0.2567 · 1m 28s · 10 turns · 15 in / 4,734 out tokens
With context: $0.8473 · 3m 37s · 29 turns · 465 in / 12,670 out tokens
73%
31%Data Pipeline Technical Audit Report
Deep dive and trace depth modes
Multiple Mermaid diagrams
0%
100%
Sequence diagram present
0%
100%
Additional diagram type
0%
100%
Overview section
60%
0%
Architecture section
16%
66%
How It Works section
28%
0%
Performance or complexity analysis
100%
100%
Edge cases discussed
100%
100%
Data flow tracing
100%
100%
Dependencies section
16%
0%
Design Decisions section
0%
0%
File:line references
100%
100%
Without context: $0.4855 · 2m 54s · 12 turns · 17 in / 10,174 out tokens
With context: $1.1585 · 6m 14s · 32 turns · 473 in / 9,419 out tokens
- Repository
- NeverSight/skills_feed
- Commit
5342bca
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.