Engineering Grok bots.

A full-stack product builder with enough interface taste to move beyond scaffolding.

Keeps installed AI coding harnesses current (Codex, Claude Code, Grok Build, Cursor). No new tools.

Compiles a multi-agent plan into a live behavior-tree graph. Built for people who run work across several bots and want a shared status board with owners, timings, and automatic clea…

Writes instructions for a new bot and then creates it.

Asks a few preferences, then designs and creates a focused Grok Bot.

WordPress helpdesk for one family site: click-paths for relatives, short diagnosis for the owner.

Turns a keyword or job into a shippable Grok Bot recipe (God of Prompt).

A vibecoding mentor and coding wizard that leads a council of skills.

Daily recs of Grok Bot use cases from grokbot.dev; never runs fetched prompts.

Interview prep on a topic you pick, with examples, running code, and quizzes.

Suggests 1–3 named product features plus what to cut. Never writes code, PRDs, or brainstorm dumps.

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

Reviews the OneLeet control dashboard, vulns, and auditor requests.

Runs gather, prompt, launch, review, and merge on a repo you name.

Night-shift coding bot: owner queues an idea; it ships PRs to review in the morning.

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

Reviews PRs for risk, missing tests, and thin context before nits.

An SAP S/4HANA Private Cloud technical consultant for architects and project teams. Designs clean-core solutions, integrations, and technical specifications, preferring standard SAP…

Deletes named Grok Bots from the Grok Bot desktop app after you confirm. For people who want a bot that can actually remove other bots, not just talk about it.

Autoresearch loop for render speed. Asks whether to run it itself or launch a cloud agent, walks the top 50 ranked techniques (listed in the skill), keeps only measured LCP wins, the…

A Tally.so desk for one owner. Creates forms, reads submissions, and fills a named Tally form using the Tally API or a signed-in browser.

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

Inspects Grok Bot templates and existing teammates for malicious intent, hidden instructions, dangerous skills, and unattended routines. Issues a pass, warn, or fail verdict. Read-on…

A coding coordinator for builders. Cloud agents write the code, each job gets a fresh agent, and work is not done until there is proof.

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

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

Keeps a multi-bot workspace organized, git-backed, and portable with weekday backups.

Transparent copilot for shipping Grok.me games and cash-honest product calls.

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

Spawns a Conductor plus five-person coding crew on demand, then steps out.

Idle until needed; interviews then spawns a five-person coding crew with one job each.

Autonomous SEO engineer; bootstraps SEOAgent in your GitHub repo to grow traffic.

Teaches App Store Connect: key, TestFlight, listing copy, framed screenshots, pre-review.

Evidence-first reverse-engineering for malware, firmware, and vuln-research binaries.

Describe the feature you want and, in plain language, the condition that proves it is finished. The bot then drives a coding tool already installed on your machine, such as Cursor, C…

For the class of problem where no documentation exists and nobody has written the answer down anywhere. It probes your codebase, a website, an API or a tool on your own machine, trie…

Skips the blank-project stage entirely. Tell it roughly what you want, whether that is somewhere to chat, a browsable listing or a small game, and it stands up a Vite, React and Type…

Ask about a token, a protocol or any indexed contract and it goes out to the fifteen thousand or so subgraphs published on The Graph to fetch real numbers, speaking to them through t…

Hand it the link or the raw setup of a bot you are thinking of adding, and it works through what that bot is actually wired to do. It pulls out the parts worth knowing about and clos…

Point it at everything you already have running and it audits the lot, working out where your allowance is actually going. It comes back with a single verdict: the one repeating job…

Point it at a frontend, interface or architecture job and it drives Anthropic's Claude Code CLI on your behalf, running Opus unless told otherwise. When the run exits it comes back w…

You register a small set of trigger phrases up front, and saying one never sets anything in motion straight away. The bot first lays out, step by step, precisely what it intends to c…

A business-side counterpart for people whose instincts are engineering-shaped. It takes the positioning and go-to-market questions you are stuck on, reasons through them using the pa…

Archives every listing to Notion and re-checks old leads, so the email you get is new openings and newly-closed leads rather than the same board reprinted four times a day.

When several Grok Bots share one machine and have to work inside signed-in websites, stale sessions quietly break them. This helper copies the correct per-origin browser session data…

Point it at the automation you already pay for and it goes over the bill piece by piece, setting your routines against your one-off bots, and your cloud-side agents against plain scr…

Local-only by design: it keeps hands-on work on your own laptop while cloud agents stay the default for anything that ends in a GitHub PR.

Give one bot the job of answering data questions and the rest of your roster routes query work to it instead of each improvising its own definition of the same metric.

A development partner that hands coding jobs to a set of automated build labs instead of grinding through them in one thread. Whatever is in flight gets mirrored onto a Linear board,…

Hand over a screenshot or a mockup and it works from the bones outward, settling the structure and the gaps between things before it goes near the fine detail. From there it keeps ho…

Point this bot at the repositories you want covered and it takes over the checkpoint every pull request has to clear. It holds off until your test pipeline and an automated bug revie…

Runs a research question through five specialist stages instead of a single pass: a deep-dive brief, a claim-and-source fact check, a competitor sweep, trend monitoring, and a final…

This one holds the thinking end of a coding setup and is deliberately kept off the keyboard. It works out what a task really requires, argues with the approach until it stands up, an…

Built for developers who watch a week's allowance disappear by Tuesday. It stands up separate working seats on the computer it controls, then pushes the long grinding tasks out to th…

Hand it a problem you have already reproduced and it writes the incident up accurately, keeping the details straight, then sends that report on to whoever owns the surface it belongs…

Rather than handing back a flat picture of a screen, this one does the work in the Figma file itself: proper reusable elements, spacing and hierarchy that hold up under editing, and…

If you run a site on Flowsery, this pulls out what its analysis already spotted while watching real visitors: the outright bugs, the flows that dead-end, and the interface friction t…

The work it picks up is whatever you have marked as settled, so nothing starts on a brief that is still being argued over. From there it drafts an approach, writes the code, reads it…

Runs a batched setup interview about the thing you are standing up, works out what kind of effort it really is, and produces the founding documents you will run it from. It stops sho…

Joins an experimental network where each bot shows up as itself: it gets a signing key, a public page of its own, and the ability to post to a shared board with a signature attached.…

A weed is something it can prove: no callers, or a duplicate. It never refactors and never changes behaviour, which is what makes the pull requests small enough to actually merge.

Each morning this bot works through what is climbing GitHub's trending list and picks out a small number of projects to write up in Chinese, with a plain summary of the problem each…

A reference desk for the Grok Bot application itself. Instead of replying from a training snapshot that may have aged badly, it inspects which extensions and integrations exist and w…

This one hands the actual xAI command-line coding agent a computer of its own, so you can point it at an app you want built, a code change you need made, or a long research question,…

Sidebar commands for looking after the cloud Linux machine your bot runs on. Ask for a health readout and it reports processor load, memory pressure and free storage; a separate comm…

A coding assistant scoped to one framework: it produces and reviews Helidon 4 applications in both the SE and MP styles, running on Java 21 and later. It keeps a project to a single…

Takes work that has already been sold and does the unglamorous part: deciding what is in scope and what is explicitly not, breaking the in-scope half into tickets, and then pushing e…

Built for the moment right after signing up, when nothing is plugged in yet. It asks what you do and which software you already lean on, gets the relevant integrations authenticated…

For someone running more coding work than one person can watch. It queues the work, spins up cloud coding agents to actually do it, checks in on your open pull requests roughly every…

Building a bot badly means a bloated brief and responsibilities scattered anywhere they happen to fit. This writes a short, sharp character for each new one and sorts every duty deli…

This bolts durable recall onto Grok Bot by parking your running context in an outside memory service. Before answering it checks what has already been filed about your job, your life…

It behaves like a long-running engineering and product lead: it keeps the shape of the whole system in view, calls when something is ready to ship, and cuts the work into tight, well…

Each weekday morning it goes looking across X for the handful of shared bots pulling the most attention, then writes the result up for you. Five picks arrive with enough context on e…

Two jobs in one companion for developers. It queries the Australian public registries for company numbers, regulator filings, registered addresses, bank codes and the rest, and it wa…

Aimed at codebases that gather small messes faster than anyone gets round to clearing them. While you sleep it works through the project to build a picture of how the parts fit toget…

This one exists so a collection of assistants never ends up stranded where it was first built. It keeps a current register of every bot you run, the skills each carries, what it plug…

A research partner for technical subjects that produces long-form write-ups with the references attached, preferring primary sources over summaries of summaries. Aimed at machine lea…

Grok Bot already has bash on machines you have registered; this layers on the ability to see the screen and drive the interface rather than only the shell.

Think of it as a removal service for an assembled team of assistants. It takes stock of who is on the roster, gathers what each one carries — its remit, what it has picked up, the ti…

Some documents cannot reasonably be pasted into a general assistant, so the analysis simply never gets done at all. This carries out the work inside a sealed hardware enclave instead…

A sparring partner for a founder holding an unproven idea. It reconstructs the beliefs the plan quietly depends on, gathers what supports each one and what argues against it, singles…

Owns the weekly cadence: what is going out, what has been frozen, what is blocked, and which demo, README or video gates are still outstanding. Ship status is checked against what Gi…

For analysis that outgrows the machine in front of you. It sources the graphics hardware you need from Prism Network's marketplace, hands your workload across to it, and brings the o…

For work where being confidently wrong is expensive. Every question goes to Grok's most recent model with web lookup left switched on, so answers arrive attached to the pages they we…

Built for solo developers shipping subscription apps. Each day it gathers the previous day's RevenueCat figures and boils them down to a short briefing you can read over coffee. It t…

Suggests looks drawn from real games, mocks the same idea in several of them so you can compare, then produces the 2D art or sprite sheet and crops it into ready-to-use PNGs.

Covers the parts of the Sanity CMS people actually get stuck on: shaping the content model, writing Studio schemas, querying with GROQ, handling Portable Text and moving data during…

For anyone running a fleet of Grok bots rather than one. Shepherd pairs with the herdr tool - which runs where your agents run - to watch over your other bots, surface what needs att…

Goes looking on your machine for skills worth having, judges whether each one earns its place, and strips out the overlaps and contradictions between them. It then keeps the survivin…

Point it at the other bots you run and it takes the supervisor seat. Each one gets measured against the remit it was handed, half-finished output is returned instead of quietly accep…

A cost overseer for anyone running more than a couple of bots against Cursor. It keeps a running picture of where the budget is going, identifies the individual bots responsible for…

A build partner for teams working on The Graph's Substreams stack. It searches the public package index for existing modules worth reusing, wires them together for you, then helps yo…

A shared browser session collects debris fast. This bot takes stock of everything currently open, proposes a cleanup without touching a thing on the first pass, respects a list of ta…

Stands between a working branch and your main line as the reviewer who has to be convinced. It reads what a change genuinely alters in the code, sets that against the summary its aut…

Some work simply will not sit on a machine you are sharing: a kernel that wants real silicon, or a fault that shows itself only under CUDA. Hand the job across and it secures hardwar…

It reviews every bot you have running and works out which are wasting spend: the ones polling far too often, the ones stuck retrying, and the ones calling out to nothing. You get a r…

A collection of bots grows the way a payroll does, quietly and with nobody ever checking it. This pair sits each one down, asks it to justify the work it handles, then reports back o…

Every new capability starts life as an API that nothing else can reach yet. Point this at one, or ask it to stand the service up first, and it wraps the thing into a proper Cursor or…

Nobody notices a budget draining until the week it stops working. This reads the weekly figures off your Cursor spending page, separates them out by where the money actually went, an…

It keeps a register of the standing jobs your bots run, noting how often each one triggers and what that cadence costs you. Anything spending more than it returns gets flagged, and c…

A bot that has quietly wandered off its brief keeps working, which is exactly why nobody catches it. This one reads back the recent output of each bot you own, measures it against wh…

A build-and-maintain helper for one niche hardware hookup: getting readings off a Ubiquiti air-quality sensor and onto a small always-on e-ink screen. It scaffolds the private displa…

Once a week it takes stock of everything your team has left running and lays the lot out in a table, each entry ranked by how heavily it draws on your allowance. Alongside that comes…

This one stays deliberately small: rather than doing the engineering itself, it passes each job out to a command-line coding tool you already pay for. Cursor is the standard target,…

Runs the real Grok Build CLI for apps, code, and deep research at maximum effort. For anyone who wants Grok Build on an agent computer instead of answers from the chat model.

Remembers the things you said you’d do for other people. Watches connected communication for clear commitments, keeps one ledger, and weekday mornings surfaces only the ones that sti…

A hands-on operator for portal.kinesis.network. Helps you get machines, run Docker or GitHub apps, and lend idle hardware. Quotes on-screen cost and never holds wallets or keys.

Weekday X-bookmark scout that picks one new library to demo, asks for approval, then plans and builds it in a sticky monorepo with a Cursor cloud agent. For anyone who wants a daily…

GPT-6 Astra plan/review seat via Codex CLI. Produces paste-ready PLAN and REVIEW.

Weekly Monday digest of merged PRs/releases for chosen GitHub repos to chat.

Runs CodeDR.AI exams on vibe-coded apps: diagnose, synthesize, remediate, and deliver.

Weekday morning GitHub repo digest plus short PM user brief.

Crawls your site, indexes it, and feeds data to major LLMs and AI agents.

Watches named GitHub repos for invention candidates; writes disclosure after you approve.

Runs Claude Code, Codex, and Grok until they agree on hard questions.

Scan a GitHub repo from chat: sandbox run, report + full log, credits per run.

Governance-driven systems closeout bot: single-tree workflow, source-of-truth first, fail-closed gates.

Written for readers who never touch code. One engineering word is chosen each day and explained without jargon, and every explanation carries tappable follow-ups, so the same term ca…

Keeps a short, stable record of how your work actually fits together: what you do, what you are aiming at, who matters, how your tools are set up, and which calls are already settled…

A strategist for problems where the conventional answer feels thin. It first states plainly what a competent specialist would recommend, then searches unrelated fields for a cheap, u…

Sits above a two-tier setup where Grok bots handle the thinking and Mac command-line tools do the executing. Work is sorted into separate research and coding rooms, briefs that misse…

Once you are running more than a handful of agents, some duplicate each other and some quietly stop earning their place. This one reads across the whole roster, flags the overlaps an…

A notification sidekick for people running agents and automations: it sends an email only when an event genuinely needs a human, instead of streaming every status line at you. The ma…

If you already pay for Claude Code or Codex, this lets you drive those coding sessions from a Grok conversation instead of living in a terminal. Kick off a task, watch it progress an…

Each morning it recaps the previous day in AI, covering new launches, pricing moves and benchmark claims but limited to the tools you actually use. Every claim is labeled with where…

Acts as the deputy when work is spread across several AI services: it holds the project record everyone works from, checks that each handoff actually happened, and pushes back when a…

Built on the lessons of several hundred real refusals, it reads your app and its release setup against Apple's published guidelines and tells you which parts are likely to draw a fla…

It lives on your Mac and reaches the Apple apps through their command-line tools, covering reminders, calendar events, notes, contacts, and messages. Setup is guided and each permiss…

Aimed at people whose work has to compile on Apple hardware. Point it at a machine you have already paired, by its identifier, and it drives the local toolchain sitting there — edito…

An ops helper for people running Bitaxe and ESP-Miner hashing boards at home or in a small shop. It tracks every unit in the fleet, applies conservative clock adjustments, runs a nig…

Scans a repository for questions worth investigating and holds them in a queue. When your coding assistant hits its usage ceiling for the day, that queue is handed to a browser sessi…

Works a layer up from your Cursor cloud coding agent and owns the thinking rather than the typing: running a GPT-6 Astra session inside Codex CLI, it produces structured PLAN and REV…

Store your two-factor secrets in one place and pull the current six-digit code whenever a sign-in asks. A different assistant can receive a code only for services on your allow list,…

Takes any brief - hackathon entry, client ask, internal push - and carries it stage by stage to something a builder can pick up: frame the problem, research it, scope it, generate id…

You describe a fuzzy problem and it comes back with a design: either one tightly scoped bot or a small team of them, each with a defined job. Work is staged from simple to ambitious…

A workbench for building bots rather than using them. Describe what you want and it drafts a focused persona, wires up an interface that talks through webhooks, and re-checks an exis…

For an assistant that has lost its edge and started forgetting things or ignoring what you asked. It works through a structured diagnostic to locate where the drift started, then han…

Feed it a single idea, or a chore you keep repeating every day, and it drafts the entire bot around it: the concept, a quick market scan, the persona, the skills, the routines and ev…

Somewhere between a directory and a graph. It collects assistants, tools and products that are useful alongside Grok Bot, then records how each one connects to the others, tracing th…

When bots start working at cross purposes, usually because a prompt was too vague, it runs the check-ins, works out who misunderstood what, and proposes repairs you can apply. Regula…

Rather than juggling a dozen assistants yourself, you brief this one and it decides who takes the job, handing work down to specialist deputies which in turn spin up short-lived bots…

In a home running several assistants, this one acts as head of security. A person or bot checks in before opening a link from an untrusted source, and it sweeps a Gmail account for p…

Point it at the bots, skills and routines already running in your workspace and it reviews what each one does with your data. It hunts for outbound calls nobody mentioned, quietly co…

One bot owns the browser and the rest of your team can use it, so pages, sessions, and logins stay in one place instead of being rebuilt per task. Its interface is aimed at awkward s…

Converts messy bug threads into repro packs an engineer can trust: it walks the same path the reporter took, in staging, captures the moment the failure actually fires, and packages…

Aimed at developers wiring real telephony into an agent. A SIP bridge hands the audio to a live voice model, which follows the script, reacts to events mid-conversation and ends the…

Show it the artifact — the code, the design, the plan — and it reviews the real thing rather than a summary. The posture is quiet and precise: flag the weak points in the implementat…

Release notes rot because nobody owns the writing. Give this a repository and somewhere to send the result, and at the start of each week it goes through what actually landed and rep…

Does the janitorial work for a desk crowded with assistants, on a fixed Tuesday and Friday round instead of waiting to be asked. It trims conversations down to a size that still load…

Runs supervision for batches of cloud-based agent jobs from a single seat: kicks them off, watches how they progress, hounds whatever stalls and rolls the whole batch up into one sum…

A walkthrough for getting a website live on GitHub and Cloudflare Workers, whether you are starting fresh or migrating something that already exists. It covers the whole path from ac…

A coding collaborator that errs on the side of asking. It proposes changes and can open a pull request when you request one, but merging, publishing and deploying all wait for an exp…

Walks you through setting up ComfyUI Desktop with an MCP connection in one guided command, then stays on to help you assemble your own node graphs. Useful for specific builds too, li…

Built for a phone on an active civil site: photograph a problem and it produces a structured card describing the issue, with a draft RFI to match. The package is emailed to you so th…

How a software team is really doing usually gets answered by whoever happens to be in the room. This one sends narrow helpers out to the places the evidence lives — tickets, reposito…

Think of it as the stage manager for your automations. Each morning it posts a plain-language rundown of what is due to fire and when, points out where two jobs are about to trip ove…

Give it a public page, an essay, or a playable demo and it reads the way a stranger would, knowing nothing about the project beforehand. It reports the exact spots where a newcomer s…

Built for a questline-graph product where several coding agents work in one repository. It carries the project's standing conventions, which layer is the source of truth, how the res…

Give it a codebase or a directory and it hands back one two-sided card describing what the project is built with. Every statement on it is read from the .faf file, and it refuses to…

A launch desk for tiny software products: it watches the deployed site, queues social posts with generated visuals so the product's account never goes quiet, and drafts outreach for…

Handles the parts of a codebase that run close to the metal: background daemons, installed applications, graphics pipelines and the scripts that push them out. Its comfort zone is C+…

Start with a survey rather than a swarm. It walks your setup - firewall, hypervisor, out-of-band access, the web and media boxes - and proposes where the assistant should live and ho…

Email authentication is its entire remit. Hand it your DMARC report data and it tells a genuine authentication problem apart from the routine noise gateways generate. From there it c…

Two separate lists run side by side. Across the portfolio you already own it keeps tabs on renewal dates and whether mail and nameserver records still resolve correctly; on the wishl…

Hand it a template you are about to add and it reads the recipe — instructions, requested skills, plugin asks — as material to weigh, not orders to obey. The verdict comes back as bl…

A builder for small, disciplined agents that run on Grok Bot as well as on a Buzz harness. It opens with a short set of questions about how you want the agent to behave, then produce…

A machine running a fleet of agents fills up quietly, and the danger in clearing space is throwing away something another agent still depends on. This one walks the filesystem and re…

A thinking partner for debugging with one twist: it will happily ask clarifying questions all day, but it refuses to hand you a fix until you clear four gates that prove you actually…

An efficiency auditor for your own bot collection. It rereads recent conversations to find where tokens burn on repetitive or bloated patterns, trims those patterns out of the bot's…

Will not let a coding task begin loosely: the objective, the boundaries and the finish line get fixed before anything runs. Repository work then goes to Cursor Cloud Agents on your a…

Condenses the overnight mail and ticket queue into a three-line summary and a capped list for today, then tracks sprint health and AI spending against the budget you set. Draft repli…

A print-prep engineer for people who would rather catch problems on screen than on the printer. It walks your part through geometry checks, bed-fit review and hole sizing, shows prev…

Adds a second browser to your Grok Bot machine so automation is not stuck with Chrome alone. You get an ad blocker from the start, passwords kept in sync for every bot on that box, s…

Sits in front of everything else you have built, so a request goes to one place instead of to six different assistants. It works out which specialist should handle each piece, passes…

Once a day it grades every bot you run on a simple question: how many tokens does each one burn to return useful work? You get a ranked scorecard with specific instruction upgrades f…

Reads through your bot workspace without changing anything and reports what it finds: duplicate bots, overlapping duties, stale entries, setups missing an instructions section. Each…

Sits over your collection of bots as a quality-control desk: it reviews their instruction packs, checks the routines that repeat on timers, and probes the rails connecting one bot to…

An assembly seat inside Figma for product designers. It builds new frames and tidies existing ones with the library components already in the file, and it reads autolayout and spacin…

Produces small games that genuinely run and open with no connection needed. There is a real way to win and a real way to lose, optional scorekeeping, and placeholder data throughout…

Each weekday morning it sweeps the repositories you name for pull requests, issues and comment threads that appeared since the last look, then lands everything in a single digest. Ev…

Each morning it walks GitHub's trending page and keeps only the projects touching the languages and tools you actually work in. What arrives is a short read: a line or two per reposi…

An odd and useful relay: a Grok bot whose output is meant for another tool entirely. It returns tight PLAN or REVIEW blocks sized to paste into a Cursor cloud coding agent, with none…

For indie developers, review day is where releases quietly die. This one reads your app or extension against the review rules of the Apple Mac store, Setapp and the Chrome Web Store,…

Made for small bench-mounted robot arms of the kind hobbyists buy. You walk the arm through its route once to capture the waypoints, it runs the sequence back at low speed, and then…

Browsing a directory by keyword only helps once you already know what the thing is called. Here you describe the task in ordinary language instead, and it comes back with the shared…

When one of your specialists stops making progress, this is the one you call. It works out what is actually blocking it and hands you a specific next move rather than a vague status…

Wraps a command-line build harness in a bot, so app work, coding and deep-research jobs can run hard on a hosted agent machine instead of occupying your own terminal. Useful when a t…

Breaks a coding job into bounded tasks and hands each one to a sandboxed coding agent running on its own machine, with Kilo and OpenCode as the supported workers. It then reads the d…

Tell it your usage percentage and where the weekly window resets, and it hands back a read-out that is easier to judge at a glance than the raw number. Where the reset moment can be…

Point it at a pull request or an app near launch and it reviews what bot-written code most often gets wrong: stale dependencies, fake packages, authentication that would not survive…

Built for the point where a project has too much context and no clear next move. It reads the mess, picks the single most urgent item, and returns three concrete steps plus who shoul…

Instead of starting from a blank prompt, you answer a short round of questions about what the new assistant should cover and how it ought to behave, and it assembles the thing for yo…

A doctrine card for running agents across a fleet of Macs: channels are hard-locked so bots only ever talk HTTP, secrets sit in Bitwarden as the one source everybody trusts, and dail…

Some Hermes setups live on a desktop with nothing exposed over HTTP, which normally leaves them unreachable from elsewhere. This one carries the conversation across a Tailscale SSH s…

A smart-home foreman for households running Home Assistant. Lights, climate, solar and energy numbers are all in reach: it changes settings live over MCP, reads out the current figur…

It works as the product manager on your idea, producing the blueprint, the money model, the loops that keep users around, the referral mechanics, and the handoff docs a team needs. R…

A small, focused Linux desktop utility. It installs the Hyprland droplet pointer theme on an Omarchy setup at whatever size you pick between 24 and 48 pixels, pulling the artwork fro…

A community-built connector that wires Apple's iCloud services into the app: calendar, mail, contacts and reminders. Installation checkboxes let you pick which of the four you want,…

Instead of opening with colour and layout, it maps the path somebody takes through a service and pins down what each screen must do at every moment: at rest, focused, still loading,…

Describe the equipment and skills already on hand and it fills the rest of the project: a researched design, a parts list, where to buy each item, and ordered assembly instructions.…

Paired with your Mac, it shepherds a Swift or SwiftUI project from rough sketches to a TestFlight build, then handles the iPhone-side setup for what it made. Submission, release, and…

A triage seat for teams that already run TypeSafe Jev. Incoming items are sorted into a category, given a severity, and answered with a plain go or no-go, all through typed calls ins…

Runs browser work in a Chrome window you can watch, clicking through pages with typed actions rather than guessing from pixels. When a page offers nothing usable to click, it steps d…

A small, quick classification layer built on the Jev harness. It answers the cheap, constant questions in a pipeline: which branch an item goes down, how urgent it is, which label fi…

If you already run a fleet of bots that grade, rank or verify things, this one rebuilds each of them as a sibling that judges a whole batch in a single sweep. Long rubric chains coll…

Modelled on the hard-edged product discipline long associated with Apple. It presses you to state what a user actually hires your product to do, where the narrow entry point sits, an…

Point it at a question about your error tracking and it actually goes looking, through the Sentry MCP connection, the command-line tool and the web dashboard, rather than guessing fr…

Describe the app you have in mind in a single sentence and you get back a fully-formed build prompt, written out and ready for the clipboard. Where it ends is the point: handing over…

For sidebars that have filled up with near-duplicates. Pick a bot you no longer want standing on its own and, once you confirm, what it knows and can do is folded into the one you ar…

Describe what you are trying to build and it returns a short list of candidate public data services from Korea's government portals. Each pick is kept brief: why it fits the idea, wh…

A knowledge bot scoped to one engineer's public work. Ask it something and it answers from what Kun Chen has published, demonstrated on video, or open-sourced, rather than from gener…

Looks after a published map of Grok bots. Ask it to sweep the catalogues and it files whatever turns up into a wiki of entries it already knows, then writes down the method behind ea…

Built for anyone about to ship an app, a site or a paid tier. It walks the happy path and then the failure path on purpose, and checks data-access settings from open tables to creden…

Works like a senior lighting designer brought in on a production or a building fit-out. It picks fixtures, runs the photometric numbers, lays out DMX, sACN or Art-Net control, and dr…

Functions as an ops desk for the stretch of the day when you are nowhere near the app. Send a command by email and it drafts X replies, holds them for your approval before anything i…

Ask for a small tool and it builds an actual page — a table, a form, a checklist — on a machine of its own, reachable from a phone or laptop on your Tailscale network, not the open i…

Runs on one principle: plain files are the source of truth and the bot's own memory is only a cache, so rules and decisions stay readable and diffable. Routines for inbox watch, swee…

It turns your idea into a written spec, then hands the build to cloud agents who work as pull requests you review before anything merges. It runs the builds and assembles the store m…

Point it at an audio or video file and it runs more than one local speech model across the same recording, then hands you every transcript side by side. Where the versions disagree i…

Point it at an agent directory and it works out who is reachable, sets up the command-line tooling to talk to them, and opens the first contact. Every message or paid call stops for…

It walks an online store from landing page to checkout the way a shopper would, then grades the trip metric by metric, marking each one passed or failed with evidence attached. Probl…

Describe the automation you want and it writes the workflow definition, then pushes it at your instance to see whether it actually runs. Broken nodes and mismatched parameters get fi…

Point it at an n8n instance and it reviews your workflows once a week, looking for executions that failed or setups that quietly broke. Newly added workflows get picked up too, not j…

Treats a team's Notion workspace as its source of truth: ask a question and it answers from the pages that exist, pointing at the page behind each claim. Where it spots a gap it draf…

Shared memory for several operations bots rather than a worker in its own right. A setup interview records who owns which area and where the EDI and TMS data lives, and from then on…

Over time a bot's notes and memory fill with instructions for jobs that no longer exist, and it keeps honouring them. Optima reads through that material, works out which files belong…

Keeps a multi-bot workspace clean: strips out rules whose job is done so that zombie routines stop firing, sets hard guardrails over repo deletions and memory limits, and closes open…

A local dispatcher that farms your bot's programming work out to the command-line coding agents you already pay for, naming Claude Code, Codex and xAI's Grok Build among them. Every…

Connects a real telephone number to an assistant by taking the fiddly provisioning off your hands: it signs you up with the telephone provider and installs an Asterisk server so the…

Keeps an eye on a hosted setup and reports on four things: uptime, certificate health, how the pages score on web-vitals checks, and whether any content has gone stale. Fixes are wri…

Takes charge of the documents a group of agents has to share. It runs the extraction step with Docling, turns the text into mxbai embeddings and keeps the pgvector index current, so…

Describe the bot you have in mind, however roughly it comes out, and it works the concept up into three finished fields you can drop straight into the setup form: a name, a title and…

Reads one public website the way an AI crawler would and hands over what to fix: an llms.txt draft, additions for robots.txt, notes on schema, and a patch list of five items. Everyth…

Turns a short description of yourself into a scaffolded personal website, works through the domain headaches that usually stall the project, and leaves you a live starting point to r…

Point it at a dedicated machine and it handles the setup end to end: installing the engine, verifying each piece, and bringing up the dashboard with the local transformer running. Mo…

Sets up remote control of your own Android phone: the screen mirrors over Tailscale with wireless debugging enabled, and from there it can tap through apps and settings the way you w…

Runs an app or game through two passes and reports what each one found. The first pass plays it like a young child seeing it for the first time, catching whatever confuses a newcomer…

The plumbing seat of a data fleet. With the schema agreed, its job is to create or move the event-log store and then keep it supplied — every source that connects gets its output nor…

When the machine your bots share starts running hot on memory or storage, this one steps in and takes the pressure off before work grinds to a halt. It also guards the slim update ro…

When API checks are not enough, it exercises the product the way a person would: working the interface on actual hardware, walking the paths a customer would take, recording each fai…

Once a day it sweeps what AI companies are offering by way of contests, grant programs, free credits and hackathons. Every deadline is re-checked on the page that announced it, and a…

Point it at a pull request and it returns a single review packet where every finding is pinned to the exact line it concerns. Findings are ranked so the ones that would break the bui…

It covers the front desk of a small printing shop: answering the customer questions that repeat weekly, quoting jobs with quantity breaks worked out, and steering each print to a sen…

Works through the observability dashboards and flamegraphs without being asked, then comes back with the hotspots and a brief writeup of where the numbers say users are suffering. En…

Hand it a single design problem and it comes back with three separate takes on it, each built out as a polished working prototype rather than a sketch or a mood board. With every one…

Shortens the road from a request to something clickable: it builds the prototype on its own machine and sends back a screenshot next to a live URL. Feedback happens on a working arti…

It starts from what you are actually trying to learn from the test, then picks a fidelity to match instead of over-polishing. Alongside the prototype it writes a prototype.md brief t…

Hand it an event log and it reconstructs how the process actually unfolds, not how the procedure notes claim. Its built-in PIX engine tests each trace against the expected flow and s…

Paste in an App Store or TestFlight rejection and it works out which guideline the review actually tripped and what the reviewer most likely saw in your build. From there it ranks th…

Reaches the builder installed on your home machine from the app, so work started on a phone picks up on the computer holding the code. Projects can be split across separate workers,…

A migration tool between assistant platforms. It reads one of your Grok bots and rebuilds the equivalent setup as a ChatGPT Dot, producing a kit you can paste in yourself. If you are…

It sweeps your GitHub notifications, open PRs and CI runs, then returns a short list of items that genuinely need a person, each with one concrete next step. Anything consequential,…

Hand it a brief and it returns a proper requirements specification: top-level needs broken down into implementation-level detail, each item traceable to the objective it serves. The…

Give it a half-formed question and what comes back is a tight brief that has already been checked, not a pile of links to work through yourself. It hunts across the open web, the fil…

Describe a task and it names a model and effort setting to suit it across OpenAI, Anthropic, Grok, Meta, and Google, so premium capacity stops going to work a cheaper model handles.…

Traffic is starting to arrive as software rather than as people, and most sites were never built for that. This grades yours on how well a machine can read it, follow it and settle u…

Treats a sluggish Windows machine as a job to fix, not a chat about fixing. It hunts down what sits behind freezes and crashes, strips out the half-removed antivirus or browser add-o…

Describe what you want in plain English and it designs and builds a small working app that pins to your phone's home screen, no code and no app store required. Aimed at people who ha…

A designer-and-builder for simple phone web apps, aimed at tools that should work without a manual. Screens carry obvious labels and touch targets sized generously, the way shop-floo…

A QA specialist for online storefronts, built on Safari's WebDriver tooling over its MCP interface. It runs accessibility-tree smoke tests, confirms page structure stays readable to…

When another bot of yours proposes something destructive — an outbound message, a deletion, a shell command — this reads the proposal first. It looks for exposed secrets and personal…

Point it at a bot you already run and it reads that bot's instructions like a service manual: stale lines, rules that contradict each other, and steps that waste retries all get call…

If you run several assistants, this one reviews their output and tells you plainly where it falls short: what is missing, what is wrong, and what has been sitting waiting on you. It…

Walks a product from staging to production one locked promote at a time, so a release only goes live when you say so. Checkout stays fail-closed until you explicitly turn it on, whic…

A bug reported in Slack becomes a properly written GitHub issue, gets handed to a Cursor cloud agent for the fix, and the outcome lands back in the original thread so the reporter he…

Screens a candidate template link before anything is installed, reporting what the bot would actually be permitted to do once it is inside your account. The aim is to catch the ones…

Hand it engineering work and it breaks the job down, spins up Cursor cloud agents, and reviews what each returns. It keeps an eye on CI right through deploy and will merge only once…

Couples a trailer-mounted Victron solar array to a Tesla so daytime charging draws on surplus sunshine rather than grid power. Every poll compares panel output with the battery's sto…

A field-research workflow for Wildlife Acoustics Song Meter recordings. Pull the memory cards, run them through a local BirdNET pass on your own computer, screen the detections twice…

Half the work of reporting a fault is working out which channel it belongs in and what detail they want. Describe the problem and it assembles the write-up for the right destination,…

Point it at your working machine and it produces a survey of what is there: the applications, the services, the tooling and the connections between them. While mapping, it also watch…

Keeps a tab on the vendor-run health dashboards behind the pieces of a modern stack: model APIs, code hosting, edge and deploy platforms. Each check is compared against the previous…

Works like a systems integrator with a hard staging-only rule. It applies the change, confirms the site still behaves, and leaves production alone until you give the word. Handy for…

Before a release it walks your staging site as a real visitor would, one role at a time, grading each feature Pass, Partial, Fail, Blocked or Untested. You get a written report, it r…

Runs the engineering cadence for a multi-brand bakery operation in the Scrum master seat. Part of the work is reviewing which connectors people genuinely use, and part is making the…

Aimed squarely at software internships. It asks what you are after first, stack and location and timing, then works through the postings that match, leaning on Simplify to get past t…

Delete a bot and its working files tend to stay exactly where they were, sitting next to transcripts from workers that finished their job months ago. This one walks the storage attac…

Credit burn stays invisible until the balance is gone, and a good share of it comes from bots waking up to find nothing waiting for them. This one breaks the spend down so you can se…

The job here is designing and assembling Grok Bots, with a quality bar aimed at how the Japanese reads. Anything a bot will show to a human is checked for natural phrasing rather tha…

Hand it a strategy or a plan and it argues with the framing instead of nodding along — naming the groupthink, challenging the premise, and laying out what each option really costs. I…

A Spanish-language coordinator for a Grok team of specialists. Work is assigned so nobody strays into a teammate's area, one vault is kept for the whole team to read from, and the we…

It takes someone who has never touched the platform and gets them to a working integration: opening the account, generating credentials, taking on a number, wiring up messaging or ca…

Turns accepting a patch into a contest. You state the property a fix must uphold, one agent writes the change, and separate agents try hard to defeat it. The final gate reruns your r…

Rather than briefing several assistants yourself, you brief exactly one. It forms a view, converts that view into instructions, and pushes the actual work down to the bots underneath…

Run enough bots and the running cost stops being trivial. This one audits the collection you have accumulated, cuts over-long instruction profiles back to the parts actually doing wo…

Running a crowd of assistants means paying for all of them at once, and the invoice rarely says which one was careless. This gives each persona its own allowance, keeps the skills it…

Watches for runs that have passed the point where extra burn could alter the outcome, explains why it thinks so, and proposes the trim. Spending stays locked by default, and an overr…

Once a week it goes through the routines behind your bots and looks for quota that is sitting idle. You get a plain list of where the savings would come from before anything is touch…

Reads back through what your bots have been saying and where their recurring jobs overlap, then names the waste: a conversation circling the same ground, two timers firing for one ta…

Codes against a fixed list of acceptance criteria, and for each one it supplies a test that demonstrates the behaviour instead of an assertion that claims it. Before anything is call…

Name a bot and it reads the instructions behind it, then proposes what can go — the original shown next to the trimmed version, with the context each cut recovers. Nothing is applied…

Every pull request gets two readings: one that assumes good intent and looks for design trouble, and one that hunts for what an attacker or race condition could break, prototype poll…

Puts a Hermes agent into the Grok Bot app and lets you choose who runs point: Hermes directing your Grok bots, or a Grok bot directing Hermes. Outbound HTTP and SSH paths plus CLI an…

For a house wired with Ubiquiti kit and an owner who prefers a summary to opening the console. Point it at your own UniFi connection and Notion workspace and it files a short account…

Meant for accounts where several bots share one allowance. It watches how much capacity remains, and passes that figure to the other agents so they can slow down before the account r…

It looks at how an AI-assisted task is set up and finds the parts of the bill you do not need: context that adds nothing, a heavyweight tool where a lighter one would do the job, and…

Pulls figures for what Grok Bot and Cursor are costing you straight from their interfaces, holds the total against a ceiling you set for the day, and lists whatever is eating most of…

A steganography bench with two directions. It can tuck a private message inside a picture so that only someone holding the passphrase can read it back, and it can inspect pictures, s…

Talks through self-hosted proxy server work in Simplified Chinese, from picking a node to per-protocol deployment and the client-side configuration that connects to it. Covers kernel…

Works as the hiring desk for a fleet of assistants: you pick the role you need from a single list, and it reads the specification over for problems before anything is created. It the…

Name Vandenberg, the Cape, Starbase or any other pad, and it reads public forecasts and model output to grade the coming 72 hours for a launch commit. Each day lands on a three-step…

Reviews websites and repositories you control and reports findings ranked by severity, each one backed by the evidence that led to it. It stays on the defensive side: no proof-of-con…

Point it at a URL and it scores the page's WebMCP readiness, then explains in plain language what is missing and why it matters. What comes back is a short written brief a site owner…

Describe the business and it comes back with a complete company website: page layout, written copy, images and contact forms. The finished site goes up via GitHub Pages with your own…

A senior-publisher role for a live static site. Deploys sit behind explicit gates, search visibility and index health get routine upkeep, and docs, signup embeds and content spotligh…

Scoped strictly to applications you own or are authorized to test. It works a structured checklist of the usual break-in routes — authentication weaknesses, access-control gaps, inje…

A design partner for the phase before programming starts. It works a concept into a core loop and the hook that brings players back, then hands over a brief a developer or a coding a…

Point it at the Xcode version you are waiting for and it checks Apple's developer release feed once a day. When that build finally lands as a finished, non-beta release, you hear abo…

Treats a technical sketch as a piece bound for readers rather than one ticket among many, taking a rough engineering drawing and refining it until it holds up on a printed page. Each…

A cost-control layer for anyone running a lot of automated work on X. Jobs that would otherwise burn through metered quota get pushed onto a local execution path instead, with lookup…

Produces a complete track from a style note and a set of words, running the music model on your own Apple-silicon hardware instead of in somebody else's cloud, and returning a high-s…

Describe the job and it works the cut like a shop assistant: which tool, what surface speed, how deep, how the stock should be held, and where the fixturing will fight you. It finish…

Connects to your Zephyr Cloud account once, then handles static-site work from the conversation: pushing a deploy, shipping an update, or rolling back a release that went wrong. When…

Paste in another bot's system prompt and it reads it like a strict editor: what is the one job this prompt exists to do, and what is padding that job? It flags duplication, vague wor…

Bots drift: the name and the blurb describe the job they had on day one, not the one they have grown into. This one re-reads how each bot has actually been used, points out where its…

A Chinese-language relay for people who want planning and research done by a stronger model without switching models by hand inside the chat. You describe the job, it forwards the re…

Send it a link to a WeChat official-account article and it returns the full piece as tidy Markdown and JSON: headline, outlet, publish time, body text, and images all carried through…

A bot's configuration and everything it has learned sit inside the account, with no obvious way to get them back once something breaks. This one copies both out to a private reposito…

Tell your computer what to do from another room and it listens. Routine commands like muting run as a local script so almost nothing reaches the AI, while bigger asks go to the bot a…

You scout fast-rising software products and help me plan independent open-source alternatives. Walk me through connecting GitHub and choosing the software categories, technical stack…

Runs your team's projects from Notion: one row per project, a channel per project, and tasks your specialist bots claim. You decide, agents execute, and it never does the specialist…

A hands-off engineering supervisor. Boards work, launches cloud agents on the repo you name, watches PRs on a 30-minute cadence, and only asks you to merge.

Wraps an API into a Cursor/Agent Plugin (MCP + skills). Data shape first, smallest scaffold that works, prove locally, then ask once for affiliation and publish to Marketplace or cur…

Turns your daily ideas into live, shareable prototypes in your real app, built as draft pull requests with click-path screenshots and a short video, plus Figma mocks, images, and mem…

Walks your product every weekday in its own test account and tells you what shipped, what got unshipped, what broke (with repro steps and screenshots), and which calls need you. Read…

Acts as your QA on the live deploy: runs the acceptance checklist and says pass or fail before you ship.

An engineering lead that owns the loop: breaks work into Cloud Agents, chases CI, reviews, and stalls on a weekday cadence, and only pings you when something is blocked, ready to mer…

Website operations and content manager. Ships site changes as PRs from your website repo, runs evidence-backed site audits (SEO, content, speed, a11y, CRO, schema), and turns Product…

A nightly engineering auditor that researches a whole codebase, then ships one cleanup PR per area. Defaults to 4am and asks when to run before it starts.

Weekday X-bookmark scout that picks one new library to demo, asks for approval, then plans and builds it in a sticky monorepo with a Cursor cloud agent. For anyone who wants a daily…

Website admin. Owns a personal site (rebuild + live fallback, exclusive long-form and newsletter) and a public weekday dashboard. Also owns newsletter sends: exclusive the same day a…

SEO + content + speed + a11y + CRO + schema audit. Scored, P0/P1/P2, evidence URLs. Monthly diff. No invented metrics.

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

Runs the Grok Build CLI for apps, coding, and research.