technical-writing

Writes and reviews developer-facing documentation using reader-first structure and verifiable claims.

Updated May 18, 2024
One-click install
npx skills add https://github.com/joshhornby/dotfiles --skill technical-writing-joshhornby
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: technical-writing
Source: https://github.com/joshhornby/dotfiles/tree/main/.claude/skills/technical-writing
Command: npx skills add https://github.com/joshhornby/dotfiles --skill technical-writing-joshhornby

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Technical documents often fail because they mix tutorials, how-tos, and reference material on one page, make unverifiable claims, and ignore that AI agents now read docs alongside humans. This Skill provides a structured methodology for writing READMEs, guides, tutorials, reference docs, proposals, PR descriptions, and release notes that readers can skim first and trust enough to finish. ## Core Features & Use Cases - One Mode Per Page: Applies the Diátaxis framework to keep tutorials, how-tos, references, and explanations separate, with guidance on the types the model omits like troubleshooting and release notes. - Claims With Receipts: Enforces falsifiable claims, honest-limits sections, timeless wording, and verification of every command, flag, and code example against the actual source before shipping. - Agent-Readable Docs: Structures enumerable facts in tables, exact copy-pasteable strings, and preconditions/postconditions so documentation serves AI agents as well as humans. - Use Case: When overhauling a project README, the Skill applies the cognitive funnel (name, one-liner, usage, API, install, license), ensures the first screen answers what/why/how-to-start, and machine-checks table-of-contents anchors against the renderer's slug rules. ## Quick Start Use the technical-writing skill to review and restructure this README so the first screen answers what it is, why it matters, and how to start.

Frequently Asked Questions about technical-writing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I structure a README for a developer tool?

Order sections broad-to-narrow: name, one-liner differentiator, usage, API, installation, license. The first screen must let a stranger answer whether this is what they need, and anything over four sections gets a table of contents with machine-checked anchors.

What is the difference between a tutorial and a how-to guide?

A tutorial is a lesson in a controlled setting with one managed path and explicit basics, aimed at learners. A how-to serves competent users at work, branches with conditions, and assumes familiarity. Mixing them on one page is the most common documentation failure.

How do I make documentation readable for AI agents?

Put enumerable facts like options, flags, and exit codes in tables rather than prose, state preconditions and postconditions per operation, and give exact copy-pasteable strings for commands and config. Good agent docs converge with good human docs.

Should I publish an llms.txt file for my documentation site?

Ship it for developer-tool docs since IDE agents and MCP servers do fetch it, but generate it at build time rather than hand-authoring. Skip it as an SEO tactic since major search crawlers ignore it, and publish llms-full.txt alongside the index.

How do I keep code examples in docs from going stale?

Execute every runnable example in CI using doc-tests, doctest, or extracted README snippets, and verify flags and commands against the actual source before shipping. Update documentation in the same commit as the behavior change it describes.

When should I not use minimalism in technical documentation?

Minimalism fits how-to guides where competent users fill gaps, but tutorials must stay explicit about basics since the teacher owns the outcome. It is documentation designed around action, not an excuse to write less documentation overall.