claw

Send prompts to NanoClaw agent containers via a Python CLI.

5|Updated Apr 2, 2026

Search Tags:#automation #devops #cli #prompts #container #nanoclaw #claw

Ready to run

npx skills add https://github.com/thmtz/nanoclaw-fleet --skill claw-thmtz

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: claw
Source: https://github.com/thmtz/nanoclaw-fleet/tree/main/.claude/skills/claw
Command: npx skills add https://github.com/thmtz/nanoclaw-fleet --skill claw-thmtz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

CLI-based access to manage NanoClaw agent containers, enabling prompt delivery without a chat app.

Core Features & Use Cases

Send prompts to NanoClaw agent containers from the terminal
Target groups by name, folder, or JID; resume sessions; pipe prompts from stdin
Auto-detects container runtime and prints the agent's response to stdout for scripting

Quick Start

Run claw to interact with NanoClaw agent containers 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 send prompts to agent containers from the terminal without a chat app?▼

You can send prompts to agent containers from the terminal using a Python CLI that delivers messages directly to NanoClaw deployments. It bypasses chat applications by routing prompts and printing agent responses to stdout for immediate scripting use.

Can I pipe prompts from stdin to automate NanoClaw agent interactions?▼

Yes, you can pipe prompts from stdin to automate NanoClaw agent interactions. The CLI supports reading prompts directly from standard input, allowing you to seamlessly integrate automated prompt delivery into your existing shell scripts and workflows.

How do I target specific agent groups by name, folder, or JID in a CLI?▼

Targeting specific agent groups by name, folder, or JID is handled directly through CLI arguments. You can specify the desired group identifier when executing the command to ensure the prompt reaches the correct NanoClaw agent container.

Does the claw CLI auto-detect the container runtime for seamless scripting?▼

Yes, the claw CLI auto-detects the container runtime to ensure seamless scripting. It automatically identifies the available runtime environment, removing the need for manual configuration when executing local prompts against NanoClaw agent containers.

What are the requirements to run a Python CLI for NanoClaw agent containers?▼

Running a Python CLI for NanoClaw agent containers requires Python 3.8 or later, an active NanoClaw deployment with a tagged agent image, and access to the NanoClaw database. These components ensure reliable local execution and accurate error reporting.

How do I resume previous sessions with NanoClaw agents from the command line?▼

You can resume previous sessions with NanoClaw agents from the command line using the CLI's session resume feature. This allows you to continue interactions with a targeted agent container exactly where the previous conversation left off.