PromptsOpenClaw
Humanizer for anything a person will read
Use case
Long replies start to sound like a model: the same hinges, the same rhythm, the same three-part lists. A humanizer pass rewrites user-facing prose before it goes out, and a small test file catches the phrases that keep coming back.
What to connect
Nothing external. A skill file plus a regression list in the workspace.
Setup
- Save the skill as humanizer.
- Create [HUMANIZER TESTS PATH] with a handful of stiff paragraphs and the phrase that should be gone after a rewrite.
- Run the tests once and keep any phrase that survives.
- Wire the skill to run on longer prose you will actually send: posts, Slack notes to other people, and anything over a short alert.
Edit before you send: How long a message must be before the pass runs, and the path of the regression file.
Prompt to paste into OpenClaw
Save this as a skill named humanizer. Run it on user-facing prose longer than [400] characters before you send it. Skip short operational alerts, backup failures, and approval prompts. Those should stay plain. Rewrite the draft so it reads like a person wrote it for this audience. Remove: - Stock hinges and filler, including phrases that announce the point instead of making it. - Stacks of dashes used as a tic. - A list of exactly three items when two or four would be honest. - Hedging piled on hedging. - Warmth that performs sincerity without adding a fact. Keep the facts, the names, and the ask. Do not add a personal story I did not give you. Regression file: [HUMANIZER TESTS PATH]. Each case is an input paragraph and a phrase that must not appear in the output. When I catch a pattern in a real message, add a case. Run the file after you change this skill. If a case fails, the skill is not done. Show me the rewrite when I ask. Otherwise send the rewritten text and do not also send the stiff draft.
More OpenClaw prompts
- Personal CRM from mail and calendar
A year of email and meetings is a pile of names you cannot search. OpenClaw reads Gmail and Google Calendar, drops newsletters and cold pitches, and keeps a local profile for each person worth remembering. You ask in plain language who you know at a company, or who has gone quiet.
- Meeting action items with an approval queue
Meetings end and the promises disappear. OpenClaw waits until a meeting is over, pulls the Fathom transcript, matches the room to your CRM, and sends you each action item to approve. Approved items become Todoist tasks. Things other people owe you stay on a waiting-on list, except people on your own team.
- Urgent email, only when it can wait no longer
You do not live in the inbox, including on weekends. OpenClaw checks mail on a timer, ignores senders you have already marked as noise, and pings a Telegram topic only when something needs you now. When you say an alert was wrong, the next check is stricter.
- Knowledge base you fill from a Telegram topic
Links you meant to remember are scattered across chats. You drop a URL into one Telegram topic. OpenClaw saves the article, the video transcript, the full post thread, or the PDF, then you ask later in plain language. A new save can point at something you already kept.