system-documentation

A system for creating and maintaining repository documentation across entrypoints, current-state guides, architecture decisions, feature specifications, tutorials, and operational notes.

In plain words
What is it for?
Use it to plan, audit, synchronize, and review documentation as the codebase changes.
Why use it?
It prevents the README from becoming an overloaded document and keeps each kind of information in an appropriate, reviewable place.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/stefaniuk/loadout/system-documentation
Any agent
npx skills add stefaniuk/loadout --skill system-documentation
Clone the repo
git clone --depth 1 https://github.com/stefaniuk/loadout

Made for: Claude Code, Codex.

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,110 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.04110
Opus 5 $0.00024 $0.02055
Sonnet 5 $0.00010 $0.00822
Haiku 4.5 $0.00005 $0.00411

Measured yesterday against content hash 116b3b79f4c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

system-documentation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.github/skills/system-documentation/SKILL.md · 391 lines

How it starts

The opening of the file, as written. The whole thing — 391 lines — stays where its author put it; the contents beside it link to each section on GitHub.

System Documentation Skill

This skill manages the repository documentation system as a set of canonical artefacts instead of treating README.md as a catch-all file.

It keeps four layers distinct:

  1. Repository entrypoints such as README.md and GitHub-conventional files
  2. Durable current-state documentation under docs/
  3. Long-lived architectural decisions under docs/adr/
  4. Feature-scoped intent and history under specs/NNN-feature-name/

The skill is opinionated on purpose. Its job is to keep documentation precise, placed correctly, and aligned with implemented behaviour.

Theoretical basis

The skill combines two established frameworks with repository evidence from mature open-source projects (Django, FastAPI, Kubernetes, Backstage):

  • Diátaxis (diataxis.fr) defines the four reader-needs modes - tutorials, how-to guides, reference, and explanation - and is used here as the content-quality control framework: every page must have exactly one mode.
  • Docs as Code (writethedocs.org/guide/docs-as-code) defines the operating model: plain-text under version control, reviewed in pull requests, validated automatically, evolved alongside the code it documents.

Diátaxis alone does not enumerate every artefact a real software system needs. Mature repositories also keep architecture overviews, ADRs, runbooks, troubleshooting/FAQ, security policy, contribution guidance, release notes, and upgrade guides as first-class documentation. This skill therefore covers the whole documentation system, not only the four core modes.

Why this exists

Documentation usually drifts in three ways:

  • canonical facts get copied into the wrong file
  • public-surface changes update code but not reference material
  • operational, architectural, and learning content get mixed together

This skill prevents that drift by managing documentation as a dependency- ordered system of artefacts with explicit contracts.

Read the full file on GitHub · 391 lines

Files

What ships with it

11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 391 lines · 48 tokens per session scan A 116b3b79f4c5

Subscribe to this mod's changes

system-documentation is a skill published in the GitHub repository stefaniuk/loadout (1 stars, last pushed 4d ago), licensed MIT. It adds 48 tokens to every session and 4,110 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.