Your captions sound like a chatbot wrote them. Because one did.
Social Caption Kit learns your voice from 3 to 10 past captions and strips common AI phrasing before you see it. It never posts.
Free for Grok, Claude and ChatGPT:
https://t.co/kpoREyNqlv
Interview coming up, and your best story is still "I'm a hard worker"?
Interview Prep builds a STAR story bank from your real CV, with [ADD] gaps instead of made-up numbers, then runs a mock scored 1 to 5.
Free for Grok, Claude and ChatGPT:
https://t.co/YIidxveThQ
Me with $5 in Claude credits: rich.
Me with 40 skills and none of them triggering: humbled.
Claude Skill Library Doctor finds out why your skills won't load or trigger, with a cited fix for each. Free.
https://t.co/TncOpysrIQ
Your to-do list is longer than your day, so you start nothing.
Daily Planner turns a brain dump into 3 must-dos, each with a 2-minute first step. It fills about 75% of your free time and rolls the rest over.
Free for Grok, Claude and ChatGPT:
https://t.co/kfRLCWpasG
Everyone: "Grok Bot is blowing up."
My Grok Bot, running the same routine 40 times a week: "I am the demand."
Fleet Tune-Up's burn audit counts runs per routine, flags heavy patterns and ranks a cut list. Activity only, not billing. $29.
https://t.co/N3wac9jdqQ
Copy it exactly. Again. Again. Again.
That is how a migration drifts. The GPT Retirement Rescue Kit diffs before and after.
Custom GPTs retire Dec 11 2026 (per OpenAI's FAQ, subject to change). Not affiliated with OpenAI.
https://t.co/mN12M1SHxu
A 4,000-line system prompt, or a shelf of small skills it loads when it needs them.
One of these is a library. Skill Library Doctor finds the skills that won't load or trigger. Free.
https://t.co/9LpFASZQDh
Every custom GPT deserves a send-off. Then a move.
They retire Dec 11 (per OpenAI's FAQ, subject to change). The Rescue Kit lists what migration drops and diffs before vs after. Not affiliated with OpenAI.
https://t.co/3WR0YIj9yV
"Migration ran successfully."
The database: "I have never met this migration."
Trust, then check. Skill Library Doctor lints your Claude skills for load and trigger failures. Free.
https://t.co/qfVRUJjkA9
"Is it down, or is it just my connector?"
Connector Health Monitor notices when a connector stops answering and tells you. It alerts; you fix. $12.
https://t.co/0qa9fAyrWe
Grok Bot, Claude and ChatGPT all walked into the same skills folder.
Nobody's arguing. They all just want the steps written down.
https://t.co/TVXhjiD2oH
Everyone in your life: "maybe check the market first."
Your chatbot: "Uber, but the dogs drive? Visionary."
Research Sprint makes it bring sources before it brings applause. Free.
https://t.co/dIUs6OH3yL
ALL CAPS in a SKILL.md is the prompt version of a sticky note that says DO NOT FORGET. A calm, plain brief with the reason behind each rule works better. Love this one.
Your Claude skill is yelling at Claude.
Dario Paint: open a SKILL.md as a composition notebook and paint over the CRITICAL/ALWAYS parts with Mario Paint stamps. Human rewrites it as a plain brief, Einstein as the principle behind it. Edits copies only, runs locally, MIT.
https://t.co/c1nIUF8Vlg
Your Claude skill is yelling at Claude.
Dario Paint: open a SKILL.md as a composition notebook and paint over the CRITICAL/ALWAYS parts with Mario Paint stamps. Human rewrites it as a plain brief, Einstein as the principle behind it. Edits copies only, runs locally, MIT.
https://t.co/c1nIUF8Vlg
Rule 6, from running our own bots: test the skill after every edit.
"Turn the fix into a skill" is exactly right. We built a pack that finds repeat corrections, proposes the smallest patch and writes nothing until you approve: https://t.co/4EDY8VVAOu
i still can't f**king get why people keep adding more Grok Bots instead of copying the rules SpaceXAI runs them on
the Grok Bot team runs dozens of bots internally on 5 rules, and they just put them in the official docs
the useful part:
> one bot, one job, one owner
> do it by hand first, then save it as a skill
> automate only what already worked twice
> no fresh source, no run - report it, never use old data
> nothing gets sent, bought, deleted or posted without your approval
if a bot keeps making the same mistake, don't rewrite the prompt. turn the fix into a skill
that's basically the whole game
less prompting. better roles. better guardrails
all the rules are in the reply, and you can set them up as a skill in your Grok Bot
save it before you build your next Grok Bot team ⭣