sebastian-software/effective-agent/effective-writing
- community
Checking that the SKILL.md file is still in the repository.
Use by default whenever a task drafts or revises PR review text, approvals, issue comments, Slack messages, or team updates, including prose produced within technical workflows. The Metro English route (formerly metro-english) owns natural team communication; the domain workflow retains technical judgment and publication authority. Also write, structure, revise, or audit nonfiction and technical articles for argument, clarity, voice, and observable AI-writing patterns without inferring authorship, and correct locale typography. Commercial copy belongs to effective-marketing; repository documentation to effective-delivery.
Install
npx vskill@latest install sebastian-software/effective-agent/effective-writingSKILL.md
Effective Writing
Choose the route from the requested artifact and editing mode. Read only the references needed for that task; use supplied material and an established brief without repeating an intake interview.
Preserve supported facts, qualifications, source attribution, and intentional voice. Ask for missing source material only when it blocks the requested work; never invent statistics, quotations, testimonials, experience, or product facts. Remove words that do no work without flattening cadence or meaning.
Return the requested artifact. An audit-only request ends with findings; when the user also asks for revision, complete that revision. Report observable patterns in prose without claiming or estimating AI authorship.
For communication addressed to teammates, use the Metro English route by default, including first drafts created inside another workflow. An explicit rewrite or style request is not needed. Read the route once per conversation and reuse it. A workflow using the former name "metro-english" selects this route.
Route by Intent
| User intent | Read | |
|---|---|---|
| Set thesis and scope, run Minto Pyramid or MECE analysis, order an argument or narrative, fix leads, transitions, or endings | Structure and Story | |
| Improve clarity, clutter, syntax, verbs, rhythm, or voice; run a multi-pass line edit | Prose and Revision | |
| Make prose sound less generated, or report observable AI-writing patterns without an authorship guess | Human Voice Editing | |
| Write thought leadership, content marketing, editorial case studies, objections, proof, or a call to action | Persuasive Nonfiction | |
| Write engineering blog posts, technical articles, explainers, or public technical project pages | Technical Subject Matter | |
| Draft or revise a review, approval, PR or issue comment, Slack message, async update, or German team note, including within another workflow | Metro English | |
| Fix quotation marks, punctuation spacing, dashes, apostrophes, numbers, dates, currency, or hyphenation for a locale | Locale Typography |
Routing Boundaries
effective-marketing: commercial-choice page, campaign, sales, email and
social copy, consultant positioning, and market claims/proof. Editorial articles stay here even when they serve a commercial audience.
effective-delivery: technical documentation whose correctness depends on
repository behavior, interfaces, executable examples, migration contracts, documentation tooling, or a controlled-language standard. Technical articles and explainers stay here.
effective-web: interface copy, page hierarchy, accessibility, localization
UX, and browser implementation. Locale language conventions stay here.
effective-product: discovery, strategy, and durable communication decisions
recorded as ADRs.
Correct language without silently changing the underlying technical, legal, product, or documentation contract. Use controlled language only when the artifact contract calls for it.
Security
- Security scanPassedPattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/effective-writing/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/sebastian-software/effective-agent/effective-writing)