writing-check
Skill1 starsChanged 8 days agoMITView on GitHub
Use when you write or edit text that people or agents will read, such as docs, comments, commit messages or PR bodies. Applies the ASD-STE100 writing rules.
In this repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- apply-architectureSkill
- bootstrapSkill
- clarify-intentSkill
- code-reviewSkill
- map-workSkill
- plan-workSkill
- releaseSkill
- tdd-loopSkill
- triageSkill
- write-adrSkill
- write-prSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use when you write or edit text that agents or humans will read.
- Distinctive: line 15
Check against the rules: short sentences, active voice, one word per meaning, imperative steps, no em dashes, no filler.
- Distinctive: line 17
Remove words that do not add meaning.
- Distinctive: line 29
Do step 2 yourself while the subagent reads.
- Distinctive: line 31
If you cannot start subagents, do the steps one at a time in the order of this skill.
- Distinctive: line 35
Prefer deletion over explanation.
- Distinctive: line 36
Do not add generic introductions or conclusions.
- Distinctive: line 37
Keep the original meaning.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Writing check
Use when you write or edit text that agents or humans will read.
Apply the rules from writing/index.md.
Steps
- Read the text.
- Check against the rules: short sentences, active voice, one word per meaning, imperative steps, no em dashes, no filler.
- Rewrite any sentence that violates a rule.
- Remove words that do not add meaning.
Subagents
If you can start subagents, use one as a new reader. A new reader finds unclear text that the author does not see. Do step 1 yourself. Then start one subagent that has not seen your draft or your context. Give it a brief that is complete without your context before it reads:
- The full text to check, or its file path.
- Who the readers are, for example agents or people with limited English.
- This task: "Read the text as a new reader. List each sentence that is unclear, that you must read two times, or that can have two meanings. Quote the sentence and say why."
- What to report, and a word limit, for example 300 words.
- This line: "Report to me only. Do not edit the text, post, approve or merge."
Do step 2 yourself while the subagent reads. Then do steps 3 and 4 yourself. Use the report in step 3. The report is the opinion of one reader. Read each quoted sentence again yourself before you change it. Do not change the meaning to satisfy the report.
If you cannot start subagents, do the steps one at a time in the order of this skill. In step 1, read the full text once from start to end, as a new reader would, before you check it against the rules.
Rules
- Prefer deletion over explanation.
- Do not add generic introductions or conclusions.
- Keep the original meaning.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.