0xa3b/writing
Writing skills for prose, technical writing, and agent instructions.
Use when creating, editing, tightening, or reviewing files that instruct agents — CLAUDE.md, AGENTS.md, SKILL.md, agent definition files, or system-prompt fragments — including reviewing a diff to such a file for instruction quality, and before writing the prompt that dispatches a sub-agent or delegated task. Provides the house style for unambiguous, auditable instructions, the document mechanics, and guidance for one-shot sub-agent prompts. Do not use as the entry point for reviewing a session's own changes before a commit or PR, which a review workflow owns and applies this skill inside; for technical artifacts such as READMEs or documentation (technical-writing); for code comments, commit messages, or chat responses; for maintaining a terminology section's terms, aliases, or relationships; for conceptual questions about instruction files; or for implementation requests the session codes directly, even when the program is an agent; delegating implementation to a sub-agent falls under the dispatch trigger.
Use when asked to write, rewrite, tighten, shorten, or review general prose for a human reader — a chat reply, explanation, summary, announcement, email, newsletter, or blog post — or to make text sound less AI-generated, robotic, or stilted, or to cut filler and slop from text. This is the base style for all prose; skills for specific artifact classes override it for their artifacts. Do not use when the target is a technical artifact or a technical comment or reply in an issue or review thread, even if the user asks to rewrite, tighten, or improve its prose. Technical-writing owns that composition after the substance and disposition are settled; verifying, triaging, or deciding how to handle existing feedback belongs to the feedback-handling workflow. Also do not use for files that instruct agents (agent-instructions); for commit messages; for program code; or for conceptual questions about writing style.
Use when writing, editing, or reviewing technical artifacts — README files, CONTRIBUTING and setup guides, runbooks, operational procedures, migration and upgrade guides, API reference prose, architecture and design documents, release notes, long-form docstrings, pull request descriptions, or issue and ticket descriptions — including making existing text clearer or easier for non-native English readers. Do not use to assess, triage, or decide changes from unresolved review feedback. Once another workflow settles the feedback, use for drafting or rewriting technical comments in Jira and other issue trackers, pull request or merge request comments, and reviewer replies. Applies Diátaxis structure, STE-inspired wording, and Google developer style formatting. Do not use for files that instruct agents such as AGENTS.md, CLAUDE.md, or SKILL.md; for chat responses, commit messages, or code comments; or for conceptual questions about writing standards.