Skip to content
Back to skills

Lean Writing

ASecurity

Use when commit messages or prose must stay tight: cut filler, padding, preamble, AI fluff, and stories of attempts; lead with current results.

  • 3 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 19, 2026
ai-agentsgoapi

Works with

  • api

Security analysis

A100/100

Pro scans all 3 files and shows the line behind each finding

Scanned September 19, 2026

npx -y skills add tony/skills --skill lean-writing --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Lean Writing?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Lean Writing
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tony-lean-writing/badge)](https://www.skillsdirectory.com/skills/tony-lean-writing)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: lean-writing
description: >-
  Use when commit messages or prose must stay tight: cut filler, padding,
  preamble, AI fluff, and stories of attempts; lead with current results.
allowed-tools: ["Read", "Grep", "Glob"]
metadata:
  source: "plugins/lean/skills/lean-writing/SKILL.md"
---

# Lean writing

Produce tight, slop-free prose and code the first time, so nothing has
to be cleaned up later.

## Core moves

- Lead with the result. Cut preamble ("Certainly!", "Here is…") and
  postamble ("Let me know if…").
- State current truth, not the journey. No diary of what changed, what
  you tried, or how you got here — the artifact is the freshest take,
  not a log.
- Reuse before you create. Search for an existing file, component,
  helper, API, test, or doc section before adding another.
- Smallest coherent change. Keep unrelated cleanup out of it.
- Preserve references when editing. Never orphan a link, citation,
  anchor, warning, or a comment that documents an invariant or "why".
- Prefer prose and nested sections over tables. Reach for a table only
  when the data is genuinely matrix-shaped and stable.
- One command per code block; keep comments outside the fence.
- No coded rule labels (`[R1]`, `Option B`) in text a human reads.

## Calibrate to the project

Read `./AGENTS.md` and `./CLAUDE.md`. When they define a slop rubric or
a house voice, that governs — match it.

## Comments

Code comments and docstrings answer to
`references/comment-policy.md`: one or two lines, only for what
the code cannot say, rationale in the commit message. To trim comments
that already exist, use the `lean-trim-comments` skill.

## Deeper catalog

See `references/lean-rubric.md` for the full
signature list, preservation rules, and the "add a table / file / test"
decision blocks.

## While tightening, don't add slop

Replacements must be concrete and shorter than what they replace. Never
narrate your own edits ("I tightened…", "cleaned up…"). Fix the text,
not a description of the fix.


## Portability notes

- Bundled files — every relative path in this skill points at a file shipped inside this skill directory. Read them from here, not from the host's plugin tree.

Files in this skill

  • SKILL.md2.2 KB
  • references/comment-policy.md5.7 KB
  • references/lean-rubric.md2.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…