agentleFS
memi-designmemi-design/design-skills/skills/content-design/SKILL.md

content-design

Skill7 starsChanged 4 months agoMITView on GitHub

Design clear, consistent product language for navigation, actions, forms, onboarding, empty states, errors, confirmations, permissions, and trust-sensitive moments. Use when words affect comprehension, action, recovery, or confidence.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat language as part of the interaction model.

    line 8
  2. Distinctive:

    Establish terminology: Create one preferred term for each concept, record synonyms and prohibited ambiguity, and align labels across …

    line 17
  3. Distinctive:

    Prepare localization: Avoid embedded word order, idioms, concatenated fragments, unexplained abbreviations, and layouts that fail under …

    line 23
  4. Distinctive:

    Test comprehension: Use cloze tests, first-click tasks, interviews, support review, and production behavior where consequence warrants it.

    line 24
  5. Distinctive:

    Read every string in its real state and sequence.

    line 40
  6. Very specific:

    Use information-architecture for taxonomy and navigation structure, interaction-design for state behavior, and ai-interaction-design when …

    line 42

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Content Design

Treat language as part of the interaction model. Do not polish isolated strings without understanding user intent, system state, consequence, and next action.

Inputs

Gather the user goal, audience knowledge, product terminology, voice principles, interface states, legal constraints, localization needs, character constraints, and evidence from research, support, search, or analytics.

Workflow

  1. Define the communication job. State what the user needs to understand or do at this moment and what the system knows.
  2. Establish terminology. Create one preferred term for each concept, record synonyms and prohibited ambiguity, and align labels across navigation, UI, help, notifications, and support.
  3. Write in action order. Lead with the information or action needed now. Use specific verbs and concrete objects. Remove throat-clearing and internal system language.
  4. Match copy to state. Distinguish initial, loading, empty, partial, success, warning, error, blocked, expired, and offline states.
  5. Design actions and consequences. Button labels should predict the immediate result. Pair destructive or high-trust actions with consequence, scope, and reversibility.
  6. Write recovery. Explain what happened, what remains safe, what the user can do next, and where responsibility lies. Do not blame the user or expose raw implementation errors.
  7. Handle trust. Explain permissions, automation, data use, billing, and irreversible changes before commitment. Avoid coercive urgency or disguised consent.
  8. Prepare localization. Avoid embedded word order, idioms, concatenated fragments, unexplained abbreviations, and layouts that fail under text expansion or RTL.
  9. Test comprehension. Use cloze tests, first-click tasks, interviews, support review, and production behavior where consequence warrants it.

Output contract

Deliver:

  • communication objective and audience context;
  • terminology decisions;
  • state-by-state copy table;
  • action and confirmation language;
  • error and recovery patterns;
  • localization and accessibility notes;
  • unresolved legal or policy dependencies.

Verification

Read every string in its real state and sequence. Confirm that labels remain distinct, actions predict outcomes, errors provide recovery, copy does not promise an unconfirmed result, and critical meaning is not carried by color or placement alone.

Use information-architecture for taxonomy and navigation structure, interaction-design for state behavior, and ai-interaction-design when language describes generated output, uncertainty, or automation.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.