# Terminal agent

**Source:** https://promtable.com/glossary/terminal-agent

> A terminal agent is an AI coding assistant that runs in the user's terminal — Claude Code, Aider, OpenAI Codex CLI — interacting with the file system and shell rather than an IDE editor.

---
A terminal agent is an AI coding assistant that runs in the user's terminal — Claude Code, Aider, OpenAI Codex CLI — interacting with the file system and shell rather than an IDE editor.

Terminal agents in 2026 are the engineer's choice for delegated work — "run the tests, fix the failures, open a PR". The interaction model: user types a task, agent reads relevant files, plans, edits files, runs shell commands, reports. Differs from IDE agents in that there's no editor surface — diffs are previewed in the terminal, no inline autocomplete. Claude Code is the production leader for Anthropic-native; Aider is the OSS leader; OpenAI Codex (2026) is the cloud-based alternative. Best for engineers comfortable in CLI who want delegation rather than IDE autocomplete.

## When to use

- Delegated autonomous engineering tasks.
- Terminal-first engineering workflows.

## Common mistakes

- Letting terminal agents run destructive commands without confirmation.
- No log retention — losing the trace of what the agent did.

## Related terms

- [agent](https://promtable.com/glossary/agent)
- [agent-loop](https://promtable.com/glossary/agent-loop)
- [ide-agent](https://promtable.com/glossary/ide-agent)

*Last updated: 2026-06-01*
---

Original page: https://promtable.com/glossary/terminal-agent
Maintained by Promtable (https://promtable.com). Content: CC BY 4.0. Cite as "Promtable — https://promtable.com/glossary/terminal-agent".
Contact: info@vibecodingturkey.com.