agentleFS
Cogni-AI-OUCogni-AI-OU/cogni-ai-agent-skills/coding-standard-writer/SKILL.md

coding-standard-writer

Skill4 starsChanged 6 months agoMITView on GitHub

Write a coding standards document for a project using the coding styles inferred from provided file(s) or folder(s).

The file

Show its rules and text

We split this skill into the rules it gives an agent

26 rules. About 0 are common, 21 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈21

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Ingest Targets: Read the contents of the target <file-path> or <folder-path>.

    line 15
  2. Distinctive:

    Format Document: Use a "minimal" or "verbose" template structure to compose the guidelines.

    line 19
  3. Very specific:

    Output Generation: Write the result to a new file (e.g., CONTRIBUTING.md, STYLE.md) or inject it into the README.md.

    line 20
  4. Distinctive:

    Template Adaptation: Use "verbose" for comprehensive style guides and "minimal" for concise rule summaries.

    line 25
  5. Distinctive:

    Contextual Insertion: When appending to README.md, locate the most logical insertion point (e.g., at the end or under a "Contributing" …

    line 26
  6. Distinctive:

    When tasked with creating coding guidelines from existing code.

    line 30
  7. Very specific:

    When generating CONTRIBUTING.md, STYLE.md, or STYLEGUIDE.md.

    line 32
  8. Very specific:

    When the project already has an explicit, well-documented CONTRIBUTING.md or .editorconfig that should not be overwritten.

    line 36
  9. Very specific:

    When generating general project documentation unrelated to coding style (use docs-writer).

    line 38
  10. Distinctive:

    Read the source contents to infer formatting patterns.

    line 49
  11. Distinctive:

    Write to the designated output location.

    line 51
  12. Distinctive:

    If file styles are extremely chaotic, prompt the user to choose a baseline reference file before generating the standard.

    line 55
  13. Distinctive:

    If requested, generate a companion test file to enforce the newly defined standards.

    line 60
  14. Distinctive:

    Indentation: Use 4 spaces per indent (or tabs).

    line 84
  15. Distinctive:

    Braces: Use the "K&R" style (opening brace on the same line) or the "Allman" style (opening brace on a new line).

    line 86
  16. Distinctive:

    All contributions should follow these rules unless otherwise noted.

    line 120
  17. Distinctive:

    Keep functions and methods small and focused.

    line 125
  18. Distinctive:

    Avoid repeating logic; prefer shared helpers/utilities.

    line 126
  19. Distinctive:

    Remove unused variables, imports, code paths, and files.

    line 127
  20. Distinctive:

    Use descriptive names. Avoid abbreviations unless well-known.

    line 131
  21. Distinctive:

    Keep comments up-to-date as code changes.

    line 166
  22. Distinctive:

    Avoid silent failures; either return errors or log them appropriately.

    line 180
  23. Distinctive:

    Write clear commit messages:

    line 187
  24. Distinctive:

    Keep pull requests reasonably small.

    line 195
  25. Distinctive:

    Write tests for new functionality.

    line 201
  26. Distinctive:

    Prefer readable test cases over complex test abstraction.

    line 203

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill Name: coding-standard-writer

<!-- markdownlint-disable MD013 MD023 MD031 MD032 -->

Dynamically generate a coding standards document based on the existing syntax, formatting, and conventions found in provided file(s) or folder(s).

Core Process

  1. Ingest Targets: Read the contents of the target <file-path> or <folder-path>.
  2. Analyze Syntax: Parse indentation, variable naming (camelCase, snake_case), commenting styles, block nesting, and string quotation formats.
  3. Identify Majority Rules: Tally occurrences of different styles; establish the majority as the standard.
  4. Handle Inconsistencies: Report formatting inconsistencies and optionally fix them only when explicitly requested by the user; always verify changes using project-specific linting tools.
  5. Format Document: Use a "minimal" or "verbose" template structure to compose the guidelines.
  6. Output Generation: Write the result to a new file (e.g., CONTRIBUTING.md, STYLE.md) or inject it into the README.md.

Core Principles

  • Empirical Inference: Standards must be derived strictly from existing code, not assumed.
  • Template Adaptation: Use "verbose" for comprehensive style guides and "minimal" for concise rule summaries.
  • Contextual Insertion: When appending to README.md, locate the most logical insertion point (e.g., at the end or under a "Contributing" heading).

When to Use

  • When tasked with creating coding guidelines from existing code.
  • To detect and fix formatting inconsistencies across multiple files.
  • When generating CONTRIBUTING.md, STYLE.md, or STYLEGUIDE.md.

When Not to Use

  • When the project already has an explicit, well-documented CONTRIBUTING.md or .editorconfig that should not be overwritten.
  • To enforce rules that contradict the organization's global engineering standards.
  • When generating general project documentation unrelated to coding style (use docs-writer).

Common Pitfalls

  • Ignoring Edge Cases: Establishing a "majority rule" based on a single small file, leading to a standard that conflicts with the rest of the massive codebase.
  • Overwriting Existing Rules: Blindly generating a new STYLE.md that overrides a carefully curated, existing style guide without asking the user.
  • Opinionated Hallucinations: Injecting Python conventions (like PEP 8) into a JavaScript project just because the LLM prefers them, rather than strictly empirical inference.

Quick Start

  1. Identify the reference file or folder.
  2. Read the source contents to infer formatting patterns.
  3. Generate the coding standard using a minimal or verbose markdown template.
  4. Write to the designated output location.

Diagnostics and Troubleshooting

  • If file styles are extremely chaotic, prompt the user to choose a baseline reference file before generating the standard.

Best Practices

  • Fetch external style guides as supplemental reference only when explicitly requested, and do not let them override the codebase's existing majority conventions.
  • If requested, generate a companion test file to enforce the newly defined standards.

What to Avoid

  • Overwriting existing documented standards without confirmation.
  • Injecting subjective preferences that contradict the codebase's existing majority style.

Coding Standards Templates

"m", "minimal"

    ```markdown
    ## 1. Introduction
    *   **Purpose:** Briefly explain why the coding standards are being established (e.g., to improve code quality, maintainability, and team collaboration).
    *   **Scope:** Define which languages, projects, or modules this specification applies to.

    ## 2. Naming Conventions
    *   **Variables:** `camelCase`
    *   **Functions/Methods:** `PascalCase` or `camelCase`.
    *   **Classes/Structs:** `PascalCase`.
    *   **Constants:** `UPPER_SNAKE_CASE`.

    ## 3. Formatting and Style
    *   **Indentation:** Use 4 spaces per indent (or tabs).
    *   **Line Length:** Limit lines to a maximum of 80 or 120 characters.
    *   **Braces:** Use the "K&R" style (opening brace on the same line) or the "Allman" style (opening brace on a new line).
    *   **Blank Lines:** Specify how many blank lines to use for separating logical blocks of code.

    ## 4. Commenting
    *   **Docstrings/Function Comments:** Describe the function's purpose, parameters, and return values.
    *   **Inline Comments:** Explain complex or non-obvious logic.
    *   **File Headers:** Specify what information should be included in a file header, such as author, date, and file description.

    ## 5. Error Handling
    *   **General:** How to handle and log errors.
    *   **Specifics:** Which exception types to use, and what information to include in error messages.

    ## 6. Best Practices and Anti-Patterns
    *   **General:** List common anti-patterns to avoid (e.g., global variables, magic numbers).
    *   **Language-specific:** Specific recommendations based on the project's programming language.

    ## 7. Examples
    *   Provide a small code example demonstrating the correct application of the rules.
    *   Provide a small code example of an incorrect implementation and how to fix it.

    ## 8. Contribution and Enforcement
    *   Explain how the standards are to be enforced (e.g., via code reviews).
    *   Provide a guide for contributing to the standards document itself.
    ```

"v", verbose"

    ```markdown

    # Style Guide

    This document defines the style and conventions used in this project.
    All contributions should follow these rules unless otherwise noted.

    ## 1. General Code Style

    - Favor clarity over brevity.
    - Keep functions and methods small and focused.
    - Avoid repeating logic; prefer shared helpers/utilities.
    - Remove unused variables, imports, code paths, and files.

    ## 2. Naming Conventions

    Use descriptive names. Avoid abbreviations unless well-known.

    | Item            | Convention           | Example            |
    |-----------------|----------------------|--------------------|
    | Variables       | `lower_snake_case`   | `buffer_size`      |
    | Functions       | `lower_snake_case()` | `read_file()`      |
    | Constants       | `UPPER_SNAKE_CASE`   | `MAX_RETRIES`      |
    | Types/Structs   | `PascalCase`         | `FileHeader`       |
    | File Names      | `lower_snake_case`   | `file_reader.c`    |

    ## 3. Formatting Rules

    - Indentation: **4 spaces**
    - Line length: **max 100 characters**
    - Encoding: **UTF-8**, no BOM
    - End files with a newline

    ### Braces (example in C, adjust for your language)

        ```c
        if (condition) {
            do_something();
        } else {
            do_something_else();
        }
        ```

    ### Spacing

    - One space after keywords: `if (x)`, not `if(x)`
    - One blank line between top-level functions

    ## 4. Comments & Documentation

    - Explain *why*, not *what*, unless intent is unclear.
    - Keep comments up-to-date as code changes.
    - Public functions should include a short description of purpose and parameters.

    Recommended tags:

        ```text
        TODO: follow-up work
        FIXME: known incorrect behavior
        NOTE: non-obvious design decision
        ```

    ## 5. Error Handling

    - Handle error conditions explicitly.
    - Avoid silent failures; either return errors or log them appropriately.
    - Clean up resources (files, memory, handles) before returning on failure.

    ## 6. Commit & Review Practices

    ### Commits
    - One logical change per commit.
    - Write clear commit messages:

        ```text
        Short summary (max ~50 chars)
        Optional longer explanation of context and rationale.
        ```

    ### Reviews
    - Keep pull requests reasonably small.
    - Be respectful and constructive in review discussions.
    - Address requested changes or explain if you disagree.

    ## 7. Tests

    - Write tests for new functionality.
    - Tests should be deterministic (no randomness without seeding).
    - Prefer readable test cases over complex test abstraction.

    ## 8. Changes to This Guide

    Style evolves.
    Propose improvements by opening an issue or sending a patch updating this document.
    ```

References

Related Skills

  • docs-writer: You MUST load this skill when writing or updating READMEs and Architectural Decision Records.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.