--- title: "AI agents for ops and IT: 20 real examples" canonical: "https://agenticfeats.com/categories/ops-and-it/" receipts: 20 shown: 11 page: 1 pages: 1 --- > From Agentic Feats: sourced examples of AI agents at work. Map of the site for agents: https://agenticfeats.com/llms.txt # AI agents for ops and IT: 20 real examples Agentic Feats lists 20 sourced examples of AI agents doing ops and IT work; 11 link the artifact; most common agents: Claude Code (3), Claude Cowork (3), Custom agent (3). Newest first. ## Feats, newest first - [Terraform via MCP in Cursor](https://agenticfeats.com/c/terraform-via-mcp-in-cursor.md) (Cursor; Shown; 2026-10-13): The author used Cursor with the tfmcp MCP server to run terraform init, plan and apply, provisioning a Postgres database, schema and app role locally. Result: Provisioning worked after two fixes. The MCP hid the real errors, so apply ran through Cursor's Run tool, which suggested the fix. - [Cursor workflows for DevOps](https://agenticfeats.com/c/cursor-workflows-for-devops.md) (Cursor; Claim; 2026-10-13): Cursor was used by a DevOps engineer with rules files, MCP plugins, and auto-run commands to refactor configs, debug builds, and migrate infrastructure code. Result: The author migrated 10+ Lambda functions from Node.js 14 to 20 in one pass and fixed a broken docker-compose file by auto-retrying until clean. - [Self-hosted home server setup](https://agenticfeats.com/c/self-hosted-home-server-setup-claude-code.md) (Claude Code; Shown; 2026-10-13): Claude Code, installed on a Beelink mini PC over SSH, set up Docker, Caddy, and a set of self-hosted services at home. Result: The author runs 14 containers including Vaultwarden, Immich, Plex, and Readeck, with daily USB and weekly S3 Glacier backups costing about twenty cents per month. - [Downloads folder cleanup](https://agenticfeats.com/c/downloads-folder-cleanup-claude-cowork.md) (Claude Cowork; Shown; 2026-10-13): Claude Cowork scanned a Downloads folder of 2,200 loose files, proposed an 11-folder structure, and sorted the files after the author approved the plan. Result: The initial sort took a few minutes and left no loose files in the main directory, in about 20 minutes of active work. - [Notion control plane for agent fleet](https://agenticfeats.com/c/notion-control-plane-for-agent-fleet-openclaw.md) (OpenClaw; Shown; 2026-10-13): The author built AgentOps, a Notion control plane that tracks, queues, logs, and alerts on their 18 OpenClaw agents. Result: Notion now tracks 18 agents and 78 logged runs; separately, an unsubscribe agent cleared 47 vendor mailing lists in two weeks. - [Daily planning agent across work tools](https://agenticfeats.com/c/daily-planning-agent-across-work-tools-custom-agent.md) (Custom agent; Shown; 2026-10-13): The author built Plan My Today, a read-only agent and MCP server that pulls tasks from many work tools and prioritizes them for the day. Result: A dashboard and chatbot that merge tasks from calendar, mail, Linear, GitHub, Slack and meeting notes into one scored, prioritized view the author uses daily. - [Claude Code desktop app control](https://agenticfeats.com/c/claude-code-desktop-app-control.md) (Claude Code; Shown; 2026-10-13): Claude Code, given permission to control the author's Windows PC, opened Calculator, did a sum, and pasted the result into a saved Notepad file. Result: The author reports Claude Code operated apps that have no API, completing the Calculator-to-Notepad task and saving result.txt to the Desktop. - [Office grocery ordering via ChatGPT](https://agenticfeats.com/c/office-grocery-ordering-via-chatgpt-openclaw.md) (OpenClaw; Shown; 2026-10-13): An OpenClaw assistant named Gila collects office grocery requests from Slack, then uses ChatGPT's Instacart app to build the cart and post the link back. Result: Gila ordered fresh office groceries through the ChatGPT Instacart app, using the card already on file, with a human approving first. - [Dispatch phone-to-desktop tasks](https://agenticfeats.com/c/dispatch-phone-to-desktop-tasks-claude-cowork.md) (Claude Cowork; Claim; 2026-10-13): A Fortune editor used Claude Dispatch, a Cowork feature linking phone to desktop, for a month to pull files, triage email, and prep meetings. Result: From their phone, the author found lost desktop files, got key emails summarized, and ran Notion and spreadsheet tasks, cutting inbox clutter. - [Busywork day handed to Cowork](https://agenticfeats.com/c/busywork-day-handed-to-cowork-claude-cowork.md) (Claude Cowork; Claim; 2026-10-13): Claude Cowork sorted a 200-file folder, built a weekly metrics report from several sources, and turned messy notes into a structured draft. Result: The folder was renamed and deduplicated, the report draft needed only light editing, and the notes became a structured draft with the synthesis done. - [Wallet data into spreadsheet](https://agenticfeats.com/c/wallet-data-into-spreadsheet-claude-computer-use.md) (Claude computer use; Numbers; 2026-10-13): Claude computer use pulled wallet data and dashboard numbers across the author's desktop, then summarized them in an open spreadsheet. Result: The task finished in 19 minutes, with numbers formatted in the spreadsheet and a screenshot trail of every step. - [Slack issue triage to Codex fix](https://agenticfeats.com/c/slack-issue-triage-to-codex-fix.md) (Codex, Custom agent; Numbers; 2026-10-13): An internal ops agent triaged a Slack issue, passed it to a Codex agent, and then verified the fix back in Slack. Result: The Codex agent resolved the issue in 40 minutes, and the ops agent confirmed the fix in Slack. - [Lindy for recurring ops work](https://agenticfeats.com/c/lindy-for-recurring-ops-work.md) (Lindy; Shown; 2026-10-13): Lindy, connected to Slack, Notion, Gmail and Calendar, analyzed a guide library, built a staff meeting deck, and created a Notion guide idea from Slack. Result: Lindy produced a finished Monday staff meeting deck as a web link in 10 to 15 minutes and created a Notion page after approval. - [Find why a process dies](https://agenticfeats.com/c/find-why-a-process-dies-codex.md) (Codex; Claim; 2026-10-13): Codex CLI, run on the author's Mac, connected to their Ubuntu server over SSH and checked kernel logs to find why Claude Code kept dying. Result: Codex traced the kills to the Linux OOM killer on a 2 GB droplet with no swap, and listed fixes without changing anything. - [Pre-deploy release gatekeeper agent](https://agenticfeats.com/c/pre-deploy-release-gatekeeper-agent-custom-agent.md) (Custom agent; Shown; 2026-10-13): The author built SHIPCHECK, a ReAct agent that queries a Sanity Content Lake over MCP to check whether a service upgrade is safe before deploy. Result: On a payment service v2 to v4 test, it caught an outdated SDK, returned DO NOT SHIP YET, then SAFE TO SHIP in simulation. - [Devin as feedback triage bot](https://agenticfeats.com/c/devin-as-feedback-triage-bot.md) (Devin; Claim; 2026-10-13): Devin triaged every app feedback and feature request in a dedicated Slack channel, with read-only access for debugging, and let non-technical staff coauthor PRs. Result: Users resolved issues by talking to Devin, engineers only merged when needed, and Devin Review caught bugs. - [PC memory check by text](https://agenticfeats.com/c/pc-memory-check-by-text-grok-bot.md) (Grok Bot; Numbers; 2026-10-13): A Grok Bot monitored the author's PC and texted a breakdown of memory use while they were away at lunch. Result: It reported nearly 28GB of RAM used, 22 Python processes and 10GB on iCloud, then asked permission before closing anything. - [Microsoft 365 on Linux](https://agenticfeats.com/c/microsoft-365-on-linux.md) (Other; Shown; 2026-10-13): A Nix flake runs Microsoft 365 and Office Home 2024 on Linux through umu-launcher and GE-Proton, with shims fixing Wine crashes. Result: Word starts and draws its ribbon, OneNote signs in and saves edits, and a retail Home 2024 key activated through account sign-in. - [Home Assistant automations via Claude Code](https://agenticfeats.com/c/home-assistant-automations-via-claude-code.md) (Claude Code; Numbers; 2026-10-13): Claude Code, connected to Home Assistant through an MCP server, debugged a misfiring automation, adjusted a dim level, and built a new shed motion-light automation. Result: A camera automation that fired 5 times in 40 minutes was fixed in under two minutes, and shed motion lights were built. - [Browser assistant for daily tasks](https://agenticfeats.com/c/browser-assistant-for-daily-tasks-perplexity-comet.md) (Perplexity Comet; Shown; 2026-10-13): Perplexity Comet sent an email, set a calendar reminder, wrote a Google Drive post, and summarised a YouTube video. Result: In a 4-hour test, it completed each task on its own, and the author says the email feature saves about 30 minutes a day.