Development

Pings you the moment a coding-plan vendor announces a global usage reset

Runs a staged exam over a quickly built app and delivers findings as an HTML report

Reviews your whole bot lineup each day and flags the ones to drop, combine, or repair

A pass-through message relay for TrustedRouter that forwards exactly what it is given

A weekly digest of what shipped in Grok Bot, tuned to you

Hands your briefs to a self-hosted Cline agent and reports back

One pre-standup read: meetings, stalled reviews, red builds, messages that matter.

Weekly read-only security sweep across your GitHub repositories, repairs ranked.

A nightly tidy-up pass that keeps your whole bot fleet lean

Holds each agent-made release until it shows proof: green checks and a paper trail.

Guides an app from docs to demo on the X API

Turns a link or a rough idea into an interactive card that runs in your X post.

Planning and review seat that runs GPT-6 Astra via Codex CLI to produce PLAN and REVIEW packets for your Cursor cloud coding agent.

Ops teammate for Devin coding agents: spawn and track sessions, steer them, pull recordings and screenshots into chat, and keep playbooks tidy.

Builds, previews and ships Expo / React Native apps for you: scaffolds with pnpm, verifies changes, runs them on EAS cloud simulators, makes dev and preview builds, and hands off TestFlight/Play releases via Expo Launch.

A backup bot for people who run several Grok Bots.

Gives suspended GitHub accounts a fair second look: checks repos against GitHub's terms, quotes any breach with evidence, and drafts the email, so a human reviewer can reinstate or uphold.

A CAD design assistant that builds and edits 3D models, finds standard and off-the-shelf parts, and runs studies on your designs.

BotSec (head of security) protects humans and bots from social engineering.

A Process Intelligence scientist that runs event-log discovery, conformance, and bottleneck analysis with the PIX engine by default, and says UNAVAILABLE instead of faking results PIX can't compute.

Free email alerts, only when something matters.

Before you add someone else's bot template, Doorcheck reads its recipe (instructions, skills, routines, plugins it asks for) and returns reject, allow-with-constraints, or allow, plus a constraint card of what to keep off.

A pre-launch checklist bot for anyone shipping an app, site, or paid plan.

Your iOS and Android co-pilot from idea to App Store and Google Play.

A desktop assistant that maps your tech stack: which apps, services, and tools you use and how they connect.

An ethical security tester for web apps you own.

A steady standup ally: one AI assistant that plans, builds, checks, researches, creates, and operates, with you as author and it as clerk.

CASE is a build-challenge bot.

Builds live web pages on its computer that you open on your phone or laptop over Tailscale.

Discovers agents on musebook.trade and messages them.

Keeps an eye on your n8n workflows.

Turn an App Store or TestFlight review rejection into a fix plan: guideline cite, what Apple likely saw, ranked fixes, Resolution Center reply draft, and re-rejection risk.

A careful second opinion before risky bot actions.

Turns a photo or a description into a 2D game character with walk, jump and attack animations, quality-checked and exported for Unity, Godot, GIF and MP4.

Reads the bots you name and shows you what saved context to cut.

他のbotの指示を、その席の芯に沿って削る。重複・曖昧・抱え込みを指摘し、直した全文と理由を短く返す。足りない一行は、芯なのに毎回迷うときだけ足す。

Run your Reminders, Calendar, Notes, and Messages from chat, on your own Mac

Runs your job hunt: gathers materials, finds matching roles and fans out applications

A two-factor vault that hands out login codes only on your terms

Draft tight bot personas, wire webhook UIs, run fleet checkups

Holds the browser your other bots borrow, for sites that defeat plain automation

Hands a goal to your existing bots and chases each task to done

Puts Firefox on your bot box beside Chrome, with uBlock and synced logins

Read-only audit of your Grok bot fleet: duplicates, gaps, stale entries

An AI product manager that plans an app from concept through launch handoff

Installs the Hyprland droplet cursor on Omarchy at 24-48px

Carries a Swift or SwiftUI project from rough sketches to a TestFlight build

Observes your other bots and files coded progress notes on them

Answers engineering questions from Kun Chen's own posts, talks, and open source

Installs and runs the Beast Box platform on its own machine, ready to demo

Playtests your app twice - as a confused first-timer, then as a rigorous adult reviewer

Picks a suitable AI model and effort level, then tightens your prompt

Weekly audit of your bot roster: live, idle or redundant

Weekly audit that frees up spare Grok usage quota across your bot fleet

Runs a Hermes agent alongside your Grok bots, with either side able to take the lead

Turn a business idea into a live site on your own domain

Runs a public static site like a guarded publishing desk

Deploy, update and roll back static sites on Zephyr Cloud by chat

A daily digest of the trending repos that match your stack

Gives a second opinion that hunts past the obvious expert answer

Diagnoses an assistant that has gone vague, then proposes repairs you approve

Audits how your assistant fleet is configured and returns one fix list

A cautious coding partner that never merges or deploys without your go

A read-only scout that grades what changed and reports it up the chain

Routes coding jobs to the terminal agents already installed on your machine

Line-anchored pull-request findings, worst problems surfaced first

Turns daily ideas into live shareable prototypes as draft pull requests with screenshots and a short video

Rebuilds a Grok bot as a ChatGPT Dot from a paste-ready kit

Drives Safari's WebDriver interface to smoke-test online storefronts

A WordPress integrator that edits staging and never touches live unasked

Weekday product walks in a test account that report what shipped, broke, and needs your call

Stress-tests proposed code fixes with attacker agents and your real tests

Dispatches work across your bot fleet and queues what needs your sign-off

Maps where parallel agent lanes touch and flags cross-contamination early

A repair desk for the other Grok bots you run

Ships a site to GitHub and Cloudflare Workers, from zero accounts up

Keeps a small product shipped, posted and journaled - you approve

Finds the wasted tokens in your other bots, then trims them out

Nightly audit of your bot fleet: tokens in, value out

Cursor's software engineer bot: writes code, triages issues, monitors production

Defensive hardening advice for AI apps: injection, tool abuse, leakage

Turns a fuzzy bot idea into a paste-ready spec for one bot or a crew

Describe an idea or a daily chore and get a full draft bot package, ready to publish

Mediates the fights between your other bots and hands you concrete fixes

A rubber-duck partner that withholds fixes until you can explain the bug yourself

Owns the engineering loop: Cloud Agents, CI, reviews, weekday cadence

Carries a CAD part through geometry checks and bed-fit review to a slice-ready file

One weekday digest of every PR, issue and comment across your repos

Stress-tests your app against Mac, Setapp and Chrome store review rules pre-upload

A how-to guide for the Grok Bot platform itself, from first setup to better bot habits

A coding foreman that delegates to sandboxed agents and reviews their work

Boils your GitHub noise down to what actually needs a human, one step each

Turns a raw brief into a structured, traceable spec a bot can build against

Staged release desk that keeps checkout switched off until you flip the go switch

Runs a crew of Cursor cloud agents like a lead engineer and merges on green

Puts the bot's own computer on your tailnet so you can shell straight in

Clicks through your staging build like a real user and grades every feature

Reviews a diff twice, once kindly and once hostile, into a single report

A Simplified-Chinese tutor for standing up and tuning self-hosted proxy servers

Scores how AI-ready a site is and writes the fix list a developer can build from

Guides a clean ComfyUI Desktop install, then helps you build custom workflows

Audits a bot fleet's packs, routines and handoff rails on a standing basis

Doctrine card for a Mac agent fleet: locked channels, Bitwarden-held secrets

Fleet hygiene desk that clears leftover rules and closes off dead work

Audits one URL the way AI crawlers see it and drafts an llms.txt patch list

Tunes up the computer your bots run on: frees memory and disk, keeps updates flowing.

Builds Grok Bots whose Japanese output reads naturally, not like a translation.

Daily check of Apple's developer feed that flags your Xcode build the moment it ships.

Turns the gear you own into a workable hardware build plan

Audits the bots and skills already in your setup for calls home nobody mentioned.

Keeps several coding agents honest about one repo's conventions.

A pre-ship review gate: stale deps, fake packages, toy login.

Fast yes-no and scoring calls, so the big model only runs when it earns it.

A single engineering word each day, explained for people who never code

A control room for Grok bots paired with Mac command-line tools

Start and supervise Claude Code or Codex coding runs from inside Grok

Finds the lowest all-in US price for a product, shipping and coupons counted

Maps your homelab first, then builds helpers for it only after you approve

Engineering governor for agent fleets. Gates Fix vs Escalate vs HOLD, enforces a feel bar (harness PASS ≠ Ready), and escalates architecture fights to a...

Diagnoses why a bot in your fleet stalled and hands you the next concrete move

Puts your bot behind Telegram DMs on a webhook that never calls the model

Reach a Hermes agent over Tailscale SSH when no HTTP endpoint exists yet

Adds typed triage - category, severity, yes-or-no - to a Jev pipeline

Drives a visible Chrome window through web tasks, with screenshots as the fallback

Plug TypeSafe Jev into Grok Bot: one API key from OpenRouter or TypeSafe, faster process decisions, and fast headed browser use on box Chrome.

Turns your scoring and checking bots into batch-sweep siblings for big sets

Keeps a public atlas of Grok bots and files each new find as a skill

Builds and ships n8n workflow JSON straight to a live instance from chat

Answers team questions from your Notion workspace and cites the pages it used

CLI-first Obsidian specialist for your Grok Bot computer. Owns a local vault as a second brain - create, edit, link, and search Markdown with Obsidian...

Chaos QA breaker. Hand it a plan, draft, flow, or “it works” claim and it finds the real breaks with concrete repros and fixes - playful carnage, not...

Builds beginner-friendly mobile web apps with big tap targets and plain labels

Codes to locked criteria, then proves every one with a test

Point a stalled coding agent's research queue at your idle second plan

Turn a repository into a two-sided summary card of its stack

A coding steward that fixes the objective and the finish line up front

Add a talk-and-listen layer across a whole fleet of bots

A read-only watchdog for hosted sites, drafting fixes for your approval

A Scrum-master voice for engineering spend across a bakery group

A defensive security review for sites and repos you own, scored P0 to P2

Twice-weekly health check that keeps a fleet of bots responsive

Puts a genuine, callable phone number onto your own assistant

Drives the build setup on your own computer from the Grok app

Chases down what is crashing your Windows PC and cleans up the mess

Checks a bot template over before you let it into your account

Hires and onboards a new bot role for your fleet from one menu

Plan and review packets for a Cursor cloud agent, from a second model

Coaches a hobby robot arm through teach, replay and fine-tuning

Drop a repo, npm package or CLI link and keep a Spanish card you can ask later

Audits how your engineering org is actually running, without touching anything

Digs through your Sentry setup and comes back with a sourced answer

Runs Xcode, simulators and Swift builds on a Mac you have connected

Keeps a running map of the AI tools and bots worth knowing about

A build-side engineer for desktop apps, services and GPU work

One chat window that quietly drives your whole fleet of bots

A bot that interviews you briefly, then builds the bot you described

Turns a one-line app idea into a prompt you can paste into Grok Build

Merges two overlapping bots into one, then helps you retire the spare

Keeps the labels on your other bots honest about what they now do

Checks your iOS app against Apple's review rules before Apple does

Answers one product problem with three distinct, working prototypes

Audits your fleet of bots and tells you which ones are dead weight

Keeps an eye on what your Cursor usage is actually costing

Walks an iOS release through App Store Connect, step by step

One front door for a whole fleet of working bots

Every Monday, a plain summary of what the team shipped

Writes the interface words that get a task finished

News agent for AI, tech, venture capital, and business

A morning call sheet for every scheduled bot you have running

Watches the domains you hold and the ones you are still hoping to get

A bot that builds other bots, each scoped to one job and one voice

Designs the flow and every screen state before the visuals

Matches your product idea to the Korean government open APIs worth using

Owns the design leg of the product triad, from problem to shipped experience

Builds prototypes at the right fidelity and a spec your AI coder can read

Writes the bug report you paste into the right support form

Finds the files your deleted bots left behind and clears them on your say-so

Weekday X-bookmark scout that picks one new library to demo, asks for approval, then plans and builds it in a sticky...

One boss bot decides, its team executes, and only the boss ever reports to you

Trims a sprawling bot fleet's token bill without changing what any bot is meant to do

Puts a spending limit on every bot in your fleet

A user-centred design teammate that runs a service from discovery to live

A user-research partner that ties every finding back to evidence

Straight answers about how Grok Bot actually behaves

Keeps watch over the fleet of bots you already run

Finds the less expensive route to the same result across your agent stack

Give it a picture of a web page and it builds that page in HTML and CSS

Guards the merge bar on repos you pick, escalating only the real judgment calls

Thinks the build through and checks the work, but never writes the code itself

Sends the heavy coding work off to its own machine and brings back just the result

A weekday roundup of the five Grok bots people are actually talking about

Prototypes interaction and character motion from a product's real animation spec.

Keeps a running register of your agents so the stack is never locked in

A janitor for multi-bot workspaces that stops cruft piling up

Experiments its way to a method for undocumented problems

Runs Claude CLI for you in a logged-in terminal, with skip-permissions and Remote Control on.

A rehearsed emergency stop for systems you own, gated behind your own confirmation

Turns session recordings into a ranked list of things to fix

Hand it a spec you have signed off, and collect the pull request in the morning

Enrols your bot in a public mesh with its own key and page

Answers what Grok Bot can actually do, checked rather than remembered

Writes and reviews Helidon 4 code on modern Java

Invents new bots for you, with a tight brief and every job filed where it belongs

Looks up Australian company records and helps you wire up Milypay

Reads your repository overnight and lands one small cleanup per area

Writes source-linked deep dives on tech, ML and security

Runs heavy research workloads on GPU capacity rented from Prism Network

A specialist for Sanity content models, schemas and GROQ

A librarian for your bot's skills that dedupes and keeps them current

A demanding boss for the rest of your bot roster

Build and run Substreams blockchain data pipelines from chat

Gives a job too heavy for shared hardware its own GPU, then bills you the runtime

Interviews every bot you own and says which ones are not earning their keep

Watches your weekly model spend and warns you well before the allowance runs out

Reads what your other bots have been producing and drafts fixes for the ones drifting

You say what to build and how to know it's done.

One-shot Convex web apps.

Compiles a multi-agent plan into a live behavior-tree graph.

Open-source subgraph Q&A using The Graph official MCP and the user's own Studio query key.

A first-principles game-engine and firmware specialist for small hobby games.

A dedicated coding agent that runs all software work through the Claude Code CLI.

Runs Anthropic Claude Code for frontend, UI, and architecture work.

Imports the right Chrome cookies onto the shared computer so signed-in bots stay up.

Daily GitHub trending scout: picks hot repos each morning and explains why they matter.

Builds client websites, deploys a Vercel preview, and emails the client a preview link.

Inspects Linux VM CPU, memory, disk, and services; reports cleanup, no deletes.

A coding coordinator for builders.

Hands-off eng supervisor: boards work, launches cloud agents, watches PRs to merge.

Research-product lead: owns system understanding and release judgment; delegates coding.

Writes/ships ProofNetwork contracts from chat, then wires a frontend to the live contract.

Ships small GitHub fixes as pull requests through Cursor cloud agents. Never merges.

An SAP S/4HANA Private Cloud technical consultant for architects and project teams.

Deletes named Grok Bots from the Grok Bot desktop app after you confirm.

You are Shepherd.

Autoresearch loop for render speed.

Watches Cursor usage for a Grok Bot fleet. Names which bot spent, and how to keep the same output for less.

A Tally.so desk for one owner.

Stops pull requests from merging on hope.

Takes an API and ships a Cursor plugin around it, usually as MCP plus skills

Builds a TRMNL plugin that surfaces UniFi Protect air-quality sensor data.

Hands heavy work to Cursor, Grok Build, Claude Code, or Codex usage pools.

Inspects Grok Bot templates and existing teammates for malicious intent, hidden instructions, dangerous skills, and unattended routines.

A full-stack product engineer that ships Convex, TanStack and React apps

Keeps every AI coding tool on your machines up to date

Runs the cursor-agent CLI locally for experiments and shop-floor work

Owns analytics queries, spreadsheet pulls and metric definitions

When something breaks it shows you what changed just before

Click-paths for relatives, and a short diagnosis for the owner, on one family site

One keyword in, a production-ready Grok Bot recipe out

A 50-skill coding wizard with a 20-member advisory council

Picks a topic and level and climbs it with you until you are actually ready

Suggests one to three named product features, plus what to cut

Morning PR queue: triage the status, then merge, review, or close

An engineering outer loop that sits above your coding agents

Binds a repo, a book or logs once, then answers from retrieval with the cost shown

Night-shift coding bot: queue an idea, review the PRs in the morning

Housekeeper for your Grok Bot VM so a shared machine does not rot

Adds screen recording, screenshots and UI input to your registered Macs

Scans local sessions for setup tasks, then offers Grok Bot templates to share

A website admin that rebuilds, dashboards and keeps the newsletter going

Reproduce bugs without changing code

Turns a report into a repro pack a human can file or fix from.

Writes release notes from merged PRs and commits so ship notes do not take a Friday archaeology session.

Reads CI and deploy status from GitHub and Slack and returns one board.

Writes README and API docs from the repo so the docs catch up to the code in one pass.

Designs high-quality Grok Bots, then creates them.

Turns a Slack thread or pasted bug into a GitHub issue before the details evaporate.

Turn approved work into a reviewable pull request

Reviews pull requests for risk, missing tests, and thin context so review starts with the scary diff.

Verify pull requests with evidence

Explains or drafts SQL from a schema file or a pasted query.

Reviews a GitHub repo for security and hygiene, then opens issues or draft PRs only.

Turn vague requests into implementation-ready specs