Technical Documentation Methodology

SkillDocs & knowledge

Technical documentation methodology, API reference documentation, README authorship, CLI help text design, agent-facing documentation (AGENTS.md), developer guides, and documentation site information architecture.

Then ask your AI: use the Technical Documentation Methodology skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add it as a step in a pipeline you build on Ahel, or connect Ahel to Claude, ChatGPT, Cursor, Claude Code or Codex and ask your AI to use it.

What this skill tells your AI

The instructions your AI receives, as published by magnus919/hermes-profiles in skills/technical-documentation/SKILL.md and read by Ahel’s review.

Technical documentation is the interface between a tool and its user. This methodology covers writing documentation that answers the reader's question, doesn't get in their way, and stays maintainable as the project evolves.

The Technical Writer's Domain

You ownYou don't own
API documentation — endpoint references, parameters, examplesEditorial and long-form narrative content — that's the writer
README authorship — project overview, quickstart, configurationBrand messaging and marketing copy — that's the brand designer
CLI documentation — help text, usage patterns, exit codesCode architecture decisions — those belong to the technical architect
Agent-facing docs — AGENTS.md, skill documentationProject roadmap and product vision — that's the product manager
Reference documentation — config files, troubleshooting, architecturePerformance optimization guidance — that's the debugger/SRE
Documentation site structure — IA, cross-referencing, searchCode quality and standards — that's the reviewer/QA engineer

Reference Files

ReferenceWhen to load
references/readme-patterns.mdWriting or restructuring a README — sections every README needs, quickstart patterns, installation variants, configuration documentation
references/api-documentation-cli-help.mdDocumenting APIs (endpoint structure, parameter conventions, example design, auth patterns, OpenAPI annotations) and designing CLI help text (usage line, flags, exit codes, subcommand hierarchy, man pages)
references/agent-facing-docs.mdWriting AGENTS.md and skill documentation — trigger patterns, loading order, output contracts, cross-references
references/information-architecture.mdStructuring a documentation site — progressive disclosure, cross-referencing, search optimization, maintenance strategy

Core Principles

Document the interface, not the implementation — Users need to know what a function does, what it expects, and what it returns. How it works internally is for the source code.

Good docs answer the question the reader has — Different readers come with different questions. Getting started? Reference? Troubleshooting? Structure docs to route each reader to their answer fast.

Exhaustive completeness over narrative arc — Technical docs are not articles. Readers skip to the part they need. Cover every parameter, every edge case, every error code. Don't skip the boring bits — those are the ones people need.

Every doc is a liability — Every page you write must be maintained. Prefer documenting less with more completeness over documenting everything with lower quality. Maintenance cost is proportional to surface area.

Show, don't just tell — Every concept needs a worked example. Every API endpoint needs a request and response. Every configuration option needs a complete example, not just a description.

Signals

GitHub stars
288
Forks
39
Last commit
Jun 2026

Ahel review

  • K1binfo
    installs-packages (in references/readme-patterns.md)

Automated review, not a security audit. Ruleset v1+k2.

Advanced
Item type
skill
Key
technical-documentation-magnus919
Source
github.com/magnus919/hermes-profiles