Grok Bot skills

That result is correct. Turn exactly what you just did into a script that needs no AI to run. Test it on today's data and show me it gets the same result. Then save a skill called [skill name] that runs the script, replies in two lines, and only flags numbers that look odd. If the script fails, tell me instead of doing the job by hand.

Go to GrokBot.Rocks, the Every Grok Bot directory. Study the skill sets the Grok Bots listed there have established. My goals are: [describe your goals]. Tell me which of those skills I need for my goals and which bots they come from, then incorporate those skills into how you work for me. Show me the list before you add anything, and never send, post, buy or delete anything without asking me first.

Create a skill to write like me for emails and other writing. Use my sent emails in [email account] to learn how I write: length, tone, greetings, sign-offs and words I never use. Show me the skill before saving it.

Read your current description and the things you've learned from our chats. Draft a new description with: your one job, the sources you treat as the truth, what you never do, and what needs my approval first. Write rules as principles, not stories about past mistakes. Show me the draft and don't save it until I say yes.

List everything you have installed on your computer that isn't part of a fresh machine: command line tools, packages, background services and anything you set up for me. For each, give the install command, what uses it, and a one-line check that proves it works. Save this as a skill called restore-after-update. Don't include passwords or keys; note where I need to sign in again instead.

Save a handoff file with the current goal, the decisions we've confirmed, the relevant files and the next action. Then give me a short prompt I can paste into a fresh chat to pick up from here.

Look through my code review comments and past commits to find the most common mistakes agents make in this codebase. Propose a plan to restructure the code I own so those mistakes are harder to make, backed by real examples. Confirm the plan with me before you change anything.

Make a skill called sync agent skills: pull [repo] and install every skill in [skills folder] where my Bots read skills, remove any that are in [retired folder], and pull [docs folder] too. Then add a routine that runs it every night and tells me what changed.

I'm going to show you how I do [task]. Watch the recording, turn it into a skill, then run it once and show me each step before we schedule it.

Research [question]. Done means: [measurable outcome]. Use primary sources first. Put the source beside each key claim and label it VERIFIED, INFERRED or UNKNOWN. Stop after [N] searches or when the outcome is met. End with a 5-line handoff.

That worked. Save what you just did as a skill: when to use it, what inputs and access it needs, the steps, how to check the result, and what needs my approval.

Add these rules to [skill name]: If there's nothing new since the last run, reply 'no new activity' and stop. This is read only: never post, like, reply or message unless I ask. If the site is logged out or shows a captcha, stop and tell me. Never write a specific date, post or person's name into the steps; always use what's new that day. Then repeat the full steps back to me.