claw

Send terminal prompts securely into NauggieClaw agent containers via CLI.

1|2|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/jboothomas/nauggieclaw --skill claw-jboothomas
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claw
Source: https://github.com/jboothomas/nauggieclaw/tree/main/.claude/skills/claw
Command: npx skills add https://github.com/jboothomas/nauggieclaw --skill claw-jboothomas

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It keeps you from needing the Auggie chat interface by letting you send prompts directly into NauggieClaw containers from the command line, so scripting, automation, and remote workflows stay fast and secure.

Core Features & Use Cases

  • Direct CLI prompts let you target any registered group by name, folder, or JID and stream the agent response back to stdout while session IDs go to stderr.
  • Session control and piping support resuming conversations with -s, reading prompts from stdin via --pipe, and listing all registered groups with --list-groups.
  • Environmental awareness detects the host runtime (container or docker), accepts custom images or timeouts, and uses the NauggieClaw database plus .env secrets so the payload is authenticated.
  • Use Case: Automate daily standups by piping meeting notes into claw to share with a project group without leaving your shell.

Quick Start

Install the claw CLI and run claw --list-groups to verify the agent connection from the terminal.

Frequently Asked Questions about claw

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate agent interactions from the terminal instead of using a chat UI?

You can automate agent interactions by sending terminal prompts directly into agent containers using the claw CLI, which streams responses to stdout and routes session IDs to stderr for scripting workflows.

Can I pipe inputs into an agent container as part of a shell automation flow?

Yes, you can pipe inputs into agent containers by using the --pipe flag to read prompts from stdin, allowing seamless integration of agent responses into existing shell automation flows.

How do I resume a previous agent session from the command line?

You can resume a previous agent session from the command line by passing the -s flag with the session ID to the claw CLI, restoring the conversation context within the agent container.

Do I need Docker to send CLI prompts to agent containers?

You need a supported container runtime such as Docker or container to send CLI prompts, along with the NauggieClaw database, .env secrets, and the nauggieclaw-agent container image for authentication.

How do I target a specific registered group when sending terminal prompts to an agent?

You can target a specific registered group by referencing its name, folder, or JID when executing the claw CLI, or run claw --list-groups to verify available agent connections.

What is the best way to script daily standups without leaving the shell environment?

Scripting daily standups is best handled by piping meeting notes directly into the claw CLI, which sends the input into the project group's agent container and returns the response to stdout.