wise-human-writinglisted

# wise-human-writing - write for humans, not for logs Before asking any user question, read and follow the [question lifecycle](../../references/workflow-host-control.md#keep-asynchronous-questions-open). Keep asynchronous prompts open until answered; this rule does not authorize questions in autonomous or otherwise prompt-free procedures. ## Why this skill exists Messages generated straight from a work session read like the work session: exhaustive, symbol-heavy, structureless. The reader of a Jira comment or a Slack update was not in that session. They need the *story* - what happened, why, what's next - in the order a human tells it, at the altitude a human cares about. This skill is the standing contract for every outbound message: it defines the structure, the altitude, and the per-tool formatting so that anything posted on the user's behalf reads like a sharp colleague wrote it. ## Two ways this skill runs 1. **Standing reference (the main mode).** Consult these rules automatically whenever composing content that will land in front of other humans through an external tool: ticket descriptions, tracker comments, PR/MR descriptions or comment replies, doc pages, chat messages. It does not matter which tracker, code host, doc system, or chat product is in play, or whether the write happens through an MCP server, a CLI (`gh`, `glab`, `jira`, `linear`), or a raw API call - if a human will read it in another tool, these rules apply. 2. **Slash rewrit