# AI use cases for Ironclaw (nearai/ironclaw) Repo: [nearai/ironclaw](https://github.com/nearai/ironclaw) | Use cases: 3 | Page: https://dedupedai.com/use-cases/nearai-ironclaw/ IronClaw is a secure, always-on personal AI assistant that stores your data locally and encrypted. Untrusted tools run in WebAssembly sandboxes, secrets are injected at the host boundary with leak detection, and HTTP requests go only to allowlisted hosts. It runs as a background service with a WebUI, connects to Telegram and Slack, runs routines on schedules, and can build new tools for itself. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Install a personal AI assistant that keeps secrets away from its tools: https://dedupedai.com/use-cases/nearai-ironclaw/install-a-sandboxed-personal-assistant/ - Switch IronClaw to another model provider safely: https://dedupedai.com/use-cases/nearai-ironclaw/switch-ironclaw-to-another-model-provider/ - Run scheduled routines and chat with your assistant on Telegram: https://dedupedai.com/use-cases/nearai-ironclaw/run-scheduled-routines-and-telegram-chat/ ## How to install a personal AI assistant that keeps secrets away from its tools with Ironclaw Repo: [nearai/ironclaw](https://github.com/nearai/ironclaw) | Difficulty: Beginner | Time: 20 minutes | Basis: From the README Install IronClaw from a release, run the guided onboarding and open the WebUI. **Who it is for:** Someone who wants an always-on assistant but worries about tools reading their keys or leaking data. **The problem:** Many agent setups pass API keys and personal data straight to every tool they run. **The outcome:** IronClaw running as a background service with an encrypted credential store and a WebUI login link. ### Steps 1. Pick an ironclaw-v release tag and run the installer script from the Quick Start 2. Run `ironclaw onboard`, choose an LLM provider and enter its key in the hidden prompt 3. Open the WebUI link it prints 4. Use `ironclaw status` later to check the service and print the login link again ### Prompt to paste into your agent ```text Install nearai/ironclaw from the latest release on this , run ironclaw onboard with , and confirm the background service and WebUI are running. ``` ### Caveat Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nearai-ironclaw/install-a-sandboxed-personal-assistant/ ## How to switch IronClaw to another model provider safely with Ironclaw Repo: [nearai/ironclaw](https://github.com/nearai/ironclaw) | Difficulty: Beginner | Time: 10 minutes | Basis: From the README Change the provider and model after onboarding without putting the key in your shell history. **Who it is for:** A user moving from one LLM provider to another, or trying a cheaper model. **The problem:** Pasting keys into config files or command lines leaves them in history and backups. **The outcome:** IronClaw routed to the new provider, with the key stored through a hidden prompt. ### Steps 1. Check the current setup with `ironclaw models status` and `ironclaw config list` 2. Run `ironclaw models set-provider openai --model gpt-5-mini` (or your provider and model) 3. Store the key with `ironclaw config set openai.api_key`; it prompts without echoing 4. Rotate the WebUI token if needed with `ironclaw config set webui.token --rotate` ### Prompt to paste into your agent ```text Switch my nearai/ironclaw setup to using models set-provider and config set with the hidden prompt, then confirm with models status. ``` ### Caveat Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nearai-ironclaw/switch-ironclaw-to-another-model-provider/ ## How to run scheduled routines and chat with your assistant on Telegram with Ironclaw Repo: [nearai/ironclaw](https://github.com/nearai/ironclaw) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Add the Telegram or Slack channel extension and set up routines that run on a schedule. **Who it is for:** Someone who wants their assistant to check things on a schedule and reach them in a chat app. **The problem:** A web-only assistant is easy to forget, and recurring checks still need a reminder from you. **The outcome:** Routines on cron schedules or webhooks, with results and chat in Telegram or Slack. ### Steps 1. Open the Extensions page in the WebUI and install the Telegram or Slack channel 2. Finish the channel setup there; channels have no config file or CLI key 3. Create a routine with a cron schedule or webhook trigger 4. Keep endpoint allowlisting on so routines only call hosts you approve ### Prompt to paste into your agent ```text In nearai/ironclaw, install the channel from the WebUI Extensions page and create a routine that every , allowlisting only . ``` ### Caveat Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/nearai-ironclaw/run-scheduled-routines-and-telegram-chat/