#577 · Primary category: AI Coding Assistants
loopndroll
Keep Codex running forever
Project last updated:04/16/26
GitHub Stars
564
Forks
43
Contributors
1
License
Other
Why we included this project
Codex tends to stop the moment it believes a task is finished, which often means tests, lint, or typecheck were skipped. Loopndroll sits on Codex's hook system and gets a say in that stop decision. You can tell it to keep the agent going until your completion commands pass, limit it to a few extra turns, or pause and wait for a reply from you. It also forwards the latest assistant message to Telegram or Slack, so you can check progress and even redirect the work from your phone. Everything runs locally, and any Telegram or Slack connection uses your own bot or webhook, so your prompts and chat data never leave your machine. For anyone who has spent an afternoon sending follow-up prompts to get an agent to actually finish, this removes that chore.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.