CLI Reference

agntk "prompt"                    Run a one-shot task
agntk -n <name> "prompt"          Named agent (persistent memory)
agntk -n <name> -i                Interactive REPL
agntk list                        List all agents
agntk info <name>                 Show agent details
agntk delete <name>               Delete an agent's state
agntk stop <name>                 Stop a running agent
agntk clean                       Interactively remove stale agents
agntk completions <shell>         Output shell completion script
agntk install <file.md>           Install a capability into the agent's harness
agntk uninstall <path>            Remove an installed capability
agntk evaluate <file.md>          Validate a capability file

Options

Flag	Short	Description
--name	-n	Agent name (enables persistent memory)
--instructions		Custom system prompt
--interactive	-i	Interactive REPL mode
--workspace		Workspace root (default: current directory)
--max-steps		Max tool-loop steps (default: 25)
--verbose		Show full tool args and output
--quiet	-q	Text output only (no follow-up, for piping)
--version	-v	Show version
--help	-h	Show help

Agent Management

Command	Description
list	Show all agents with status (running/idle) and last active time
info <name>	Agent details — memory files, workspace, sub-agents, token usage, disk size
delete <name>	Delete an agent’s state directory (with confirmation)
stop <name>	Send SIGTERM to a running agent; SIGKILL if it doesn’t exit
clean	Interactive picker to bulk-delete idle agents
completions	Output shell completion script (bash, zsh, fish)
install	Validate and install a capability (rule, instinct, skill) into the harness
uninstall	Remove an installed capability file
evaluate	Validate a capability file without installing it

Interactive REPL

In REPL mode (-i or follow-up after one-shot), press TAB after / for autocomplete:

Command	Description
/help	Show available commands
/tools	List available tools
/agents	List all agents
/model	Show current model info
/memory	Show agent memory files
/status	Show session stats
/verbose	Toggle verbose output
/clear	Clear conversation history
/exit	Quit the REPL

Shell Completion

Tab completion for commands, flags, and agent names in your shell:

# Install globally for shell completion
npm i -g agntk

# Completions auto-install on first run (bash/zsh/fish)
# Or generate manually:
agntk completions zsh

When installed globally, agntk auto-detects your shell, writes the completion script to ~/.agntk/completions/, and patches your rc file. No manual setup needed.

Examples

# Fix bugs
npx agntk -n coder "the login page crashes when the session expires — find and fix it"

# DevOps
npx agntk -n ops --instructions "you manage k8s deploys" "roll back staging"

# Research
npx agntk "compare React Server Components vs Astro islands — pros, cons, benchmarks"

# Code review
npx agntk "review src/ for security issues and suggest fixes"

# Pipe anything
git diff | npx agntk "write a commit message for this diff"
cat package.json | npx agntk "are any of these dependencies outdated?"

# Agent management
npx agntk info coder
npx agntk delete old-agent
npx agntk clean

---