agentleFS
cenconq25cenconq25/claude-code-app-studio/.claude/skills/team-content/SKILL.md

team-content

Skill40 starsChanged 7 months agoMITView on GitHub

Orchestrate the content pipeline. Coordinates content-strategist, content-designer, localization-lead, and ux-designer to take a content brief from concept through written, localized, and shipped state.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No argument -> walk the user through a brief intake via AskUserQuestion (purpose, audience, channels, deadline, length guidelines).

    line 41
  2. Very specific:

    Read the brief, the relevant PRD, and the current voice/tone guidelines from .claude/docs/content-guidelines.md (if present).

    line 45
  3. Distinctive:

    If a string should be rewritten in source, loop back to content-designer for a revised version.

    line 121
  4. Very specific:

    Add new keys to the source string table per the /localize --extract rules.

    line 130
  5. Very specific:

    When translations come back, use /localize --validate to spot mechanical issues.

    line 140
  6. Very specific:

    Ask before writing to production/content/[brief-slug]-package.md.

    line 182
  7. Distinctive:

    If any subagent returns BLOCKED:

    line 204

≈ “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.

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

Team Content

The cross-functional pipeline that turns a one-line content need into shipped, localized, on-tone copy. Useful any time a feature spans multiple screens of copy or any time a marketing-adjacent moment needs consistent voice.


Team Composition

  • content-strategist — voice, tone, content architecture, channel planning.
  • content-designer — UX writing on screen-by-screen basis.
  • localization-lead — translation pipeline, locale-specific decisions, RTL audit, cultural review.
  • ux-designer — ensures copy fits the layout and the user journey.

Spawn each via Task. Run independent subagents in parallel where their inputs are independent.


Phase 1: Resolve the Brief

Parse arguments:

  • --brief=<path> -> read the explicit brief.
  • --feature=<name> -> glob design/prd/**/*[name]*.prd.md and related stories.
  • No argument -> walk the user through a brief intake via AskUserQuestion (purpose, audience, channels, deadline, length guidelines).

Read the brief, the relevant PRD, and the current voice/tone guidelines from .claude/docs/content-guidelines.md (if present).


Phase 2: Voice and Tone Anchor via content-strategist

Spawn content-strategist via Task. Prompt template:

Brief: [content]. Project voice/tone guide: [path or summary]. Recent comparable copy in the app (from prior PRDs, prior releases): [list]. Define the voice anchor for this piece — tone, formality, humor allowance, sensitive-topic guardrails. Identify the channels this content must work across (in-app, push, email, store listing, social).

Render anchor. Use AskUserQuestion to confirm.


Phase 3: Information Architecture (parallel with Phase 4)

Spawn ux-designer via Task. Prompt template:

Brief: [content]. Voice anchor: [reference]. Map the content to screens and surfaces. For each surface, capture: container (modal, banner, inline), max length budget, dynamic vs static, related imagery, accessibility constraints.

Render the surface map. This dictates the length envelopes content-designer must write within.


Phase 4: First Draft via content-designer

Spawn content-designer via Task. Prompt template:

Brief: [content]. Voice anchor: [reference]. Surface map: [reference]. Write the first draft for every surface. Include alternates for length-constrained surfaces (push notification at 65 chars vs 120 chars). Use placeholder names (no PII). For dynamic strings, mark ICU placeholder positions.

Render draft. Use AskUserQuestion per surface group to approve, edit, or reject.


Phase 5: UX Review via ux-designer

Spawn ux-designer via Task with the approved draft. Prompt:

Approved draft: [reference]. Verify each string fits its surface at minimum and maximum supported font scales. Verify no string truncates on the smallest target device. Flag any string that needs a layout change.

Surface any layout escalations. Iterate with content-designer if needed.


Phase 6: Localization Plan via localization-lead

Spawn localization-lead via Task. Prompt template:

Approved English draft: [reference]. Target locales: [list from technical-preferences]. Produce a translation plan: vendor route, expected turnaround, special notes per locale (RTL, formal-vs-informal address in German/French/Japanese, plural rules for Russian/Polish/Arabic). Identify any string that should be rewritten in source for clearer translation. Identify any string that should be culturally adapted (transcreated) rather than translated.

If a string should be rewritten in source, loop back to content-designer for a revised version.


Phase 7: Pre-Translation Source Lock

Once English source is final:

  • Add new keys to the source string table per the /localize --extract rules.
  • Snapshot the source via /localize --freeze if release proximity warrants it.
  • Hand off to translation vendor with the briefing produced in Phase 6.

Phase 8: Cultural / RTL Review

When translations come back, use /localize --validate to spot mechanical issues. Then spawn localization-lead once more to qualitatively review:

  • Cultural adaptation correctness.
  • RTL layouts on a device for any RTL locale.
  • Tone preserved per voice anchor.

Iterate as needed.


Phase 9: Render the Content Package

# Content Package — [brief name]

## Brief
[content]

## Voice Anchor
[from content-strategist]

## Surface Map
| Surface | Container | Length budget | Constraints |

## Strings
| Key | English | Surfaces | Notes |

## Localization Plan
- Vendors: [list]
- Locales: [list]
- Transcreation flags: [list]
- ETA: [date]

## Approvals
- Content Strategist
- Content Designer
- UX Designer
- Localization Lead

Ask before writing to production/content/[brief-slug]-package.md.


Phase 10: Update State

Append to production/session-state/active.md:

## Content Package — [date]
- Brief: [name]
- Strings: [count]
- Surfaces: [count]
- Locales: [count]
- Package: [path]
- Next: hand to vendor; resume after translations land

Error Recovery

If any subagent returns BLOCKED:

  • content-strategist blocked on missing voice guide -> propose creating one this cycle.
  • ux-designer blocked on missing surface info -> escalate to design team to provide screens.
  • localization-lead blocked on no vendor configured -> surface; either configure now or accept English-only ship.

Quality Gates / PASS-FAIL

  • PASS — every surface has a string, every string within budget, voice anchor preserved, localization plan attached, transcreations flagged where appropriate.
  • FAIL — surface without a string, length over budget, missing localization plan for a launch locale.

Examples

Example 1 — Push notification system rewrite: Brief: re-tone every push to be warmer and respect frequency caps. Surface map: 14 push templates. Voice anchor: warmer, less transactional. Drafts produced, UX confirms layout, localization plans transcreation for Japan and Brazil. Package written.

Example 2 — Empty-state copy across the app: 12 screens missing empty-state copy. Brief: encouraging, helpful. Drafts produced; one screen needs a layout change to fit the chosen phrasing. Two iterations. Final package shipped to localization.


Next Steps

  • Vendor returns translations -> /localize --validate then this skill resumes Phase 8.
  • Strings ready -> /dev-story per surface to wire into UI.
  • For ongoing copy work -> include the package as the source of truth in future PRDs.

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.