asd-ste100 Skill (danyuchn) for Claude Code: What It Does and How to Install It

Makes Claude rewrite dense or ambiguous English into short, plain sentences that an AI agent or a non-native reader cannot misread.

Installs: 8.3K; on skills.sh
Stars: 5.9K; on GitHub
Updated: Nov 2026; last commit

The skill applies the ideas of ASD-STE100, the Simplified Technical English standard from the aerospace and defense industry. It rewrites text with active voice, one instruction per sentence, no semicolons and no phrasal verbs, and keeps hedges such as "may have failed" as they are. By default it returns only the rewritten text, and it shows a rule table when you ask.

It is not for marketing or creative copy. SKILL.md says it does not reproduce ASD's approved word list, so it is a clarity tool inspired by STE and not a certified STE authoring tool. It bundles a Python linter, scripts/ste-lint.py, which SKILL.md describes as stdlib-only.

Good for

People writing tool descriptions, error messages, system prompts or inter-agent instructions
Teams whose text goes to translators or non-native English readers
Writers of READMEs, changelogs and status reports who want flatter, clearer prose

Install asd-ste100

Claude Code

Copy this prompt.

Install the asd-ste100 skill in Claude Code. Run this command:

npx skills add danyuchn/asd-ste100-skill

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

The repo is a single skill with its SKILL.md at the repo root. The README documents the skills CLI and a git clone into ~/.claude/skills/. The plugin route comes from the repo's marketplace file, which lists one plugin, asd-ste100, with no hooks or MCP config in the tree. There is no Claude app tab because the skill bundles a Python script.

From the project README. Source.

What's in the skill

When it loads

Use when English text must be parsed without a human to resolve ambiguity — tool descriptions, error messages, inter-agent instructions, system prompts, status reports — and misreading has a real cost, or when text reads as dense, hedged, or easy to misparse. Triggers: disambiguate, STE100 rewrite, apply Simplified Technical English, plain-language rewrite, controlled-language rewrite, rewrite so an agent cannot misread this. Not for creative or marketing copy.

Sections in SKILL.md

When to Use This Skill
Two Modes
Source and Scope
Core Rewrite Rules
Structural rules — apply these
Lexical rules — direction of travel only
Simple tenses — apply with one exception
Scan Checklist
Process
Output Format
Boundaries
Additional Resources

Key facts

Author: danyuchn (plugin.json); the LICENSE names Dustin Yuchen Teng as copyright holder
License: MIT (LICENSE file and plugin.json agree)
Installs on skills.sh: 8,343 (Nov 2, 2026)
GitHub stars: 5,922 (Nov 2, 2026)
Last repo commit: Nov 1, 2026
What it includes: 9 files in the repo: SKILL.md, README, LICENSE, 2 example files, 1 reference file, the linter and 2 plugin manifests
Version in frontmatter: 0.4.0
Source standard: ASD-STE100 Issue 9 (Jan 2025): 53 writing rules in 9 sections, per SKILL.md

What it includes

Applies every rule, including length caps, to procedures, errors and tool descriptions
Chosen from the text type if you do not say. The choice stays internal unless you ask for the rule table.
Try: “Rewrite this tool description so an agent cannot misread it”
Strict mode
Applies the structural rules in full and the word rules as advice, for READMEs and prose
Keeps sentence length caps, active voice and no semicolons, and drops the one-word-one-meaning lockdown.
STE-flavored mode
Active voice, no phrasal verbs, one instruction per sentence, 20 or 25 word caps
Also no semicolons, noun clusters of 3 words at most, no ellipsis and paragraphs of 6 sentences at most.
Structural rules
Six habits: synonym rotation, hedges, nominalization, marketing words, run-ons, phrasals
Scanned before any rewrite.
Scan checklist
Rewritten text only by default; a rule violation table on request
A single "Kept as-is:" line appears when a longer phrasing was kept on purpose to avoid losing precision.
Output formats
Linter for the structural rules, with --json, --baseline, --disable and --selftest
Reads the files or stdin you give it. Exits 1 when hard violations exceed the baseline; passive voice and compound tenses are advisory.
Runs code · scripts/ste-lint.py
Before-and-after examples and a rule summary with citations
examples/before-after.md, examples/linter-edge-cases.md and references/writing-rules.md.
Read · examples/ and references/

From the official docs

Permissions

Low risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It runs a bundled Python script when asked
SKILL.md tells the agent to run scripts/ste-lint.py for a first pass. It takes files or stdin and prints findings. Its --selftest writes temporary files. github.com
Install telemetry
The README says the skills CLI sends the skill name and a timestamp, and that DISABLE_TELEMETRY=1 opts out. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Scopes	Optionally runs scripts/ste-lint.py on files or stdin you give it The skills CLI install sends anonymous install telemetry

Questions

What does asd-ste100 do?

It makes Claude rewrite English into Simplified Technical English style: short, active sentences with one meaning each. It targets text an AI agent, a translation pipeline or a non-native reader has to parse without help.

Does it make my text official ASD-STE100?

No. SKILL.md says it is a general-purpose clarity tool inspired by STE and not a certified authoring tool. It does not include ASD's approved word list, so for aircraft maintenance documentation you must check against the real dictionary.

How do I install it?

Use the install box above. The README documents npx skills add danyuchn/asd-ste100-skill and a git clone into ~/.claude/skills/. The repo also has a Claude Code plugin manifest, but the README does not document it.

Should I use it on marketing copy?

No. SKILL.md says it is not for creative or marketing copy because STE is deliberately flat and literal.

Will it change what my text means?

It is built not to. SKILL.md says to keep every fact, condition and scope qualifier, keep hedges at their original strength, and flag trade-offs instead of dropping a safety condition. Still read the result.

Do I need Python?

Only for the optional linter. The rewrite works without it. SKILL.md describes ste-lint.py as stdlib-only, so it needs a Python interpreter and no extra packages.

Does the install send data anywhere?

The README says the skills CLI sends anonymous install telemetry (skill name and timestamp, no personal or device information) and that DISABLE_TELEMETRY=1 turns it off.

Is it safe?

The skill rewrites text. The optional linter is a Python script that reads files or stdin you pass to it. Our advice: run it only on files you chose, and check rewrites of safety text yourself.

Information

updated Nov 2, 2026

Repository: danyuchn/asd-ste100-skill

danyuchn/asd-ste100-skill skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 6

README.md · github.com
marketplace.json · github.com
plugin.json · github.com
SKILL.md · github.com
ste-lint.py · github.com
writing-rules.md · github.com