Find buying signals on Reddit
A reviewer mentions a Grok Bot that found buying signals on Reddit.
Real workflows with the tools, steps and prompts to copy them. Filter by what you use and how much time you have.
A reviewer mentions a Grok Bot that found buying signals on Reddit.
The poster uses Grok as a terminal coding agent that reads, edits, runs tools and spawns sub-agents.
The author runs multiple Grok Bot agents connected to Microsoft email and Teams accounts, including two BD agents and one sourcing agent.
The poster built the browser 4X game Hexwardia in two weeks using Grok and Gemini to write the combat engine, fog-of-war, map generation and heuristic AI.
The poster built a game where Grok generates the story narration, and says it outperformed Gemini and GPT for narrative output.
The poster had Grok 4.5 build a level of an existing Three.js game from a detailed game design document, followed by some manual cleanup.
The poster built a cinematic studio in Grok Imagine using three custom agents for prompt direction, audio scripting and character-consistent video.
A user runs several Grok Bot agents: one updates a Google Drive sheet of properties, one replies to clients in their email tone, and one builds a day-job command center.
Over five months, the author treated bugs like patients and organized coding agents as a hospital-style team to diagnose and fix them.
Agents are given access to social accounts to watch for comments, DMs and mentions and notify the user when one is worth a reply.
A user had Claude Cowork read 400 PDF files and sort them into different folders.
A CrewAI crew checks 50 competitor pages, changelogs and news sources every week and sends a short brief covering only what changed.
A developer used ChatGPT Agent mode to extract revenue growth, free cash flow and net cash from 10-Q filings with source URLs, feeding a multi-agent valuation and sentiment workflow.
A developer's custom agent searched forums and review sites, cross-referenced findings, and produced a cited 3,000-word competitor analysis with target users and an outbound playbook.
A hobbyist used agents to design multiple decks of an Excelsior-class starship as a 3D explorable model and to write personalized fanfic prose around it.
The author built a small Claw-like agent loop on Kimi and GLM5 that reads email read-only, messages via a self-hosted Matrix server, keeps memory files, and handles reminders and tasks.
Manus built a Google Form survey on undergraduates' views of AGI; the author hit a snag when asking it to fill the survey with synthesized student responses.
The author gives agents a skill file and docs so they generate recipe URLs for a recipe viewer, working even in permission-limited agents like ChatGPT web.