Analysis updated 2026-05-18
Chat with a personal AI assistant through a Telegram bot connected to Claude.
Run a personal automation script that starts an ongoing AI conversation.
| goroman/nullevi03 | electron/governance | dockur/proxmox | |
|---|---|---|---|
| Stars | 160 | 162 | 149 |
| Language | Shell | Shell | Shell |
| Setup difficulty | moderate | easy | — |
| Complexity | 3/5 | 1/5 | — |
| Audience | general | developer | ops devops |
Figures from each repo's GitHub metadata at analysis time.
Requires a paid Claude subscription, a Telegram bot via BotFather, and Claude Code Channels configured, README offers almost no technical detail.
Nullevi03 is a personal shell script project described by its author as an AI secretary. The README is written in Japanese and is very short, offering only a rough outline of what is involved. According to the README, using it requires a paid Claude subscription (the author mentions $200 per month as an example cost), a Telegram account with a bot set up through Telegram's BotFather service, and a working Claude Code Channels configuration. Once those are in place, you start the project using a script called boot.sh and then converse with the AI. The README offers no technical explanation of how the pieces fit together, no code examples, and no detail about what the AI secretary actually does beyond general conversation. Troubleshooting advice is a single line: ask the AI, not the author. The license is described as a custom one that does not yet exist, and the disclaimer states the author takes no responsibility for anything that goes wrong. This appears to be a small personal project shared publicly in a very informal state. Anyone looking to use or understand it would need to read the source code directly, as the documentation provides almost no guidance.
A small, informally documented personal shell script project that acts as an AI secretary using a paid Claude subscription and Telegram.
Mainly Shell. The stack also includes Shell, Claude, Telegram.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.