Practical overview
When it fits
Use klaw when the work is operating agents—creating and organizing them, scheduling tasks, and checking what they are doing—rather than building an agent framework. Slack control is relevant when operators want the documented status, run and scheduling commands inside Slack.
Who it is for
- Teams operating multiple AI agents and needing status, logs and scheduled runs in one CLI.
- Developers who want to manage agent tasks from Slack as well as from a terminal.
- Operators who need namespace-level configuration for scoped secrets and allowed tools.
Documented uses
- Create agents with a namespace, model and skills.
- Schedule recurring agent tasks, such as lead scoring or competitor monitoring.
- Inspect agent status, details and logs, or trigger tasks through CLI and Slack.
- Configure model providers and namespace-level secrets and tool permissions.
Requirements & limits
The public GitHub repository describes klaw as source-available under the each::labs License. Its README says internal business use, personal projects and consulting are free, while multi-tenant SaaS or white-label distribution requires a license. The README documents Go 1.24+ and an install script.
- The project distinguishes agent operations infrastructure from frameworks used to build agents.
- Providers may be configured using the documented provider integrations or an OpenAI-compatible endpoint with a base URL and model.
- Slack control requires Slack bot and app tokens; the CLI is documented as a separate channel.
- Namespace settings can scope secrets and allowed tools, described by the project as logical isolation.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install and configure klaw for CLI chat
Install the CLI and configure an Anthropic provider for interactive chat.
Before you start
- Shell access to run the documented install command.
- An Anthropic API key, set as ANTHROPIC_API_KEY.
Steps
Run the documented install script: curl -fsSL https://klaw.sh/install.sh | sh.
Set ANTHROPIC_API_KEY in the environment.
Run klaw chat.
Expected outcome
The documented setup leads to an interactive CLI chat using the configured Anthropic provider.
Guide limitations
- This guide covers CLI chat setup only, not creating or deploying an agent.
- The README also lists OpenRouter and each::labs as provider options.
Create and schedule a lead-scoring agent
Define an agent and an hourly task using the documented CLI commands.
Before you start
- klaw installed and a model provider configured.
- A CRM skill and web-search skill for the example agent; the task refers to HubSpot.
Steps
Create the agent in the sales namespace with a selected model and crm,web-search skills.
Create a cron job with the hourly schedule, agent name and lead-scoring task.
Use klaw get agents -n sales to inspect the namespace, and klaw logs lead-scorer to view logs.
Expected outcome
The README’s example defines an hourly lead-scoring job and shows commands for checking agent status and logs.
Guide limitations
- The README supplies an example, not evidence of a successful run.
- HubSpot credentials and the required skills must be configured for the example task.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 16 October 2026
- Original website title
- klawsh/klaw.sh · Official homepage title · retained in its source language