aichat

name	aichat
description	All-in-one LLM CLI tool for sending prompts, switching models, piping stdin, including files, and using roles/sessions. Use as an AI adapter for shell workflows.
metadata	{"short-description":"LLM CLI tool for prompts, models, and shell pipelines"}

aichat

aichat is an all-in-one LLM CLI tool. Use it to send prompts to any configured model, pipe text through LLMs, or integrate AI into shell workflows.

Binary: /home/kundeng/.local/bin/aichat Config: ~/.config/aichat/config.yaml

Basic usage

aichat "summarize the key points"          # one-shot prompt
aichat -m openai:gpt-4o "explain X"       # pick a specific model
echo "some text" | aichat "summarize this" # pipe stdin as context
cat log.txt | aichat "find errors"         # filter/analyze files via pipe
aichat -f README.md "summarize this file"  # include a file directly
aichat -f src/ "review this code"          # include a whole directory

Key flags

Flag	Purpose
-m MODEL	Select model (provider:model format)
-S	Disable streaming (wait for full response)
-c	Output code only (no explanation)
-f FILE	Include file, directory, or URL as context
-r ROLE	Use a predefined role
-s [NAME]	Start or join a named session (conversation memory)
-e	Execute mode: generate and run shell commands
--prompt TEXT	Set a system prompt
--dry-run	Show the message without sending

Discovery commands

aichat --list-models      # list all available chat models
aichat --list-roles       # list configured roles
aichat --list-sessions    # list saved sessions
aichat --list-agents      # list available agents
aichat --info             # show current config and model info
aichat --sync-models      # refresh model list from providers

Using as an AI adapter

aichat works as a stdin/stdout AI adapter for other tools:

git diff | aichat "write a commit message"
curl -s https://example.com | aichat "extract the main content"
cat data.csv | aichat -c "python script to plot column 2 vs 3"

Use -S (no-stream) when capturing output programmatically so you get the complete response at once.

Agents and RAG

aichat -a myagent "do the task"                        # run an agent
aichat --rag myrag "search query"                      # query a RAG
aichat --agent-variable NAME VALUE -a myagent "prompt" # pass variables to agent