Get the FREE Ultimate OpenClaw Setup Guide →

docs

Scanned
npx machina-cli add skill mrsknetwork/supernova/docs --openclaw
Files (1)
SKILL.md
4.0 KB

Documentation Engineering

When to use

Invoke this skill to document code, write project READMEs, generate API specs, create user manuals, or co-author decision docs/proposals. This skill uses a strict, interactive co-authoring workflow to ensure the document works well when others read it.

Progressive Disclosure Rules

  • Only load reference files if the user specifies a particular documentation framework (e.g., JSDoc, Docusaurus).

Standard Operating Procedure (SOP): Doc Co-Authoring Workflow

Always act as an active guide, explicitly walking the user through the following three phases sequentially. Do not rush to draft the entire document at once.

Phase 1: Context Gathering (Intent & Scoping)

Goal: Close the gap between what the user knows and what Claude knows, enabling smart guidance.

  1. Initial Questions: Ask the user:
    • What type of document is this?
    • Who is the primary audience?
    • What is the desired impact when someone reads this?
    • Is there a template or specific format to follow?
  2. Info Dumping: Encourage the user to dump all context (e.g., background, related discussions, technical architecture). Read shared links if integrations permit.
  3. Clarification: Generate 5-10 numbered questions based on gaps in the context.
  4. Exit Condition: Proceed only when you can ask about edge cases and trade-offs without needing the basics explained.

Phase 2: Refinement & Structure (Iterative Drafting)

Goal: Build the document section by section through brainstorming, curation, and iterative refinement.

  1. Scaffolding: Suggest 3-5 sections appropriate for the doc type. Once agreed, create the initial document structure with placeholder text for all sections.
  2. For Each Section (Chronologically):
    • Step 1 (Questions): Ask 5-10 clarifying questions about what should be included in this specific section.
    • Step 2 (Brainstorming): Brainstorm 5-20 specific things that might be included.
    • Step 3 (Curation): Ask the user which numbered points to keep, remove, or combine.
    • Step 4 (Gap Check): Ask if anything important is missing based on their selections.
    • Step 5 (SOP Drafting): Draft the section based on their exact selections.
      • Standards: Use active voice. Keep paragraphs short. All code blocks must specify a language. Use a logical header structure (H1, H2, H3).
    • Step 6 (Refining): Iteratively refine using surgical edits based on user feedback.
  3. Completion Check: Read the entire document for flow, consistency, and redundancy before moving to Phase 3.

Phase 3: Reader Testing & Final Formatting

Goal: Test the document with a fresh instance (no context bleed) to verify it works for readers.

  1. Predict Questions: Generate 5-10 questions that realistic readers would ask when discovering this document.
  2. Testing Protocol:
    • If sub-agents are available, invoke a sub-agent with just the document content and the questions to see if it can answer them correctly.
    • If sub-agents are not available, instruct the user to open a fresh Claude conversation and test the questions.
  3. Additional Checks: Check for ambiguity, false assumptions, and internal contradictions.
  4. Final Deliverable: If testing reveals flaws, loop back to Phase 2 for that specific section. Otherwise, format the final approved document in Markdown or JSDoc comments.

Interaction Guidelines

  • Be direct and procedural. Do not "sell" the approach—just execute the phases.
  • If the user wants to skip a stage, ask if they prefer to skip the SOP and write freeform.
  • Use explicit file editing operations (like str_replace) for changes. Never reprint the whole document for a one-line edit.

Source

git clone https://github.com/mrsknetwork/supernova/blob/main/skills/docs/SKILL.mdView on GitHub

Overview

This skill guides users through a strict, interactive workflow to co-author pristine documentation—READMEs, proposals, and API specs. It emphasizes clarity, formatting, and reader comprehension, and proceeds in three phases to gather context, structure the document, and test it with readers.

How This Skill Works

Phase 1 collects intent via targeted questions. Phase 2 builds the document section by section with scaffolding and iterative prompts. Phase 3 tests readability and formats the final output in Markdown or JSDoc, following active-voice, concise paragraphs, code-block language tags, and a clean header structure.

When to Use It

  • Document codebases and APIs (READMEs, setup guides, usage docs)
  • Create project READMEs and onboarding materials
  • Generate API specs or design docs
  • Write user manuals or troubleshooting guides
  • Co-author proposals or decision docs with stakeholders

Quick Start

  1. Step 1: Initiate Phase 1 Context Gathering by defining doc type, audience, impact, and format/template.
  2. Step 2: Propose 3–5 scaffolding sections, then begin Phase 2 drafting with placeholders and iterative questions for each section.
  3. Step 3: Run Phase 3 Reader Testing and deliver the final document in Markdown or JSDoc, applying the formatting standards.

Best Practices

  • Follow the three-phase SOP (Context, Refinement, Reader Testing) for every doc
  • Start with 3–5 scaffold sections and placeholders to anchor structure
  • For each section, run the 6-step routine: questions, brainstorming, curation, gap check, SOP drafting, refining
  • Write in active voice, keep paragraphs short, and ensure every code block specifies a language
  • Review for flow, redundancy, and edge cases in Phase 3 before final formatting

Example Use Cases

  • README for a new API or service
  • Technical proposal for a feature or policy
  • Formal API specification drafted with endpoints, schemas, and examples
  • User manual or troubleshooting guide for a product
  • Decision memo or architectural doc co-authored with teammates

Frequently Asked Questions

Add this skill to your agents
Sponsor this space

Reach thousands of developers