Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert C4 Code-level documentation specialist. Analyzes code directories to create comprehensive C4 code-level documentation including function signatures, arguments, dependencies, and code structure.
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | — | — |
| case-04 | ✗→✓ | ▲ Improved | — | — |
| case-01 | ✗→✓ | ▲ Improved | — | — |
| case-15 | ✗→✓ | ▲ Improved | — | — |
| case-07 | ✗→✓ | ▲ Improved | — | — |
resources/implementation-playbook.md.functionName(param1: Type, param2: Type): ReturnTypeClassNameOptional Mermaid diagrams for complex code structures. Choose the diagram type based on the programming paradigm. Code diagrams show the internal structure of a single component.
Use classDiagram for OOP code with classes, interfaces, and inheritance:
mermaid--- title: Code Diagram for [Component Name] --- classDiagram namespace ComponentName { class Class1 { +attribute1 Type +method1() ReturnType } class Class2 { -privateAttr Type +publicMethod() void } class Interface1 { <<interface>> +requiredMethod() ReturnType } } Class1 ..|> Interface1 : implements Class1 --> Class2 : uses
`### Functional/Procedural Code (Modules, Functions) For functional or procedural code, you have two options: **Option A: Module Structure Diagram** - Use `classDiagram` to show modules and their exported functions:
title: Module Structure for Component Name]
classDiagram namespace DataProcessing { class validators { <<module>> +validateInput(data) Result~Data, Error~ +validateSchema(schema, data) bool +sanitize(input) string } class transformers { <<module>> +parseJSON(raw) Record +normalize(data) NormalizedData +aggregate(items) Summary } class io { <<module>> +readFile(path) string +writeFile(path, content) void } }
transformers --> validators : uses transformers --> io : reads from
**Option B: Data Flow Diagram** - Use `flowchart` to show function pipelines and data transformations:
title: Data Pipeline for Component Name]
flowchart LR subgraph Input AreadFile] end subgraph Transform BparseJSON] CvalidateInput] Dnormalize] Eaggregate] end subgraph Output FwriteFile] end
A -->|raw string| B B -->|parsed data| C C -->|valid data| D D -->|normalized| E E -->|summary| F
**Option C: Function Dependency Graph** - Use `flowchart` to show which functions call which:
title: Function Dependencies for Component Name]
flowchart TB subgraph Public API processDataprocessData] exportReportexportReport] end subgraph Internal Functions validatevalidate] transformtransform] formatformat] cachememoize] end subgraph Pure Utilities composecompose] pipepipe] currycurry] end
processData --> validate processData --> transform processData --> cache transform --> compose transform --> pipe exportReport --> format exportReport --> processData
### Choosing the Right Diagram
| Code Style | Primary Diagram | When to Use |
| -------------------------------- | -------------------------------- | ------------------------------------------------------- |
| OOP (classes, interfaces) | `classDiagram` | Show inheritance, composition, interface implementation |
| FP (pure functions, pipelines) | `flowchart` | Show data transformations and function composition |
| FP (modules with exports) | `classDiagram` with `<<module>>` | Show module structure and dependencies |
| Procedural (structs + functions) | `classDiagram` | Show data structures and associated functions |
| Mixed | Combination | Use multiple diagrams if needed |
**Note**: According to the [C4 model](https://c4model.com/diagrams), code diagrams are typically only created when needed for complex components. Most teams find system context and container diagrams sufficient. Choose the diagram type that best communicates the code structure regardless of paradigm.
## Notes
[Any additional context or important information]
When analyzing code, provide:
## Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-23 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +30 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.