kollabor-harness

Customize and troubleshoot the Kollab terminal AI harness with Python plugins and JSON lifecycle hooks.

1|Updated May 5, 2026
One-click install
npx skills add https://github.com/kollaborai/kollab --skill kollabor-harness
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kollabor-harness
Source: https://github.com/kollaborai/kollab/tree/main/bundles/skills/kollabor-harness
Command: npx skills add https://github.com/kollaborai/kollab --skill kollabor-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the knowledge gap for users who want to customize, extend, or troubleshoot their Kollab terminal AI workspace, which has a complex event-driven architecture, flexible configuration system, and extensive plugin ecosystem that is not immediately obvious to new users.

Core Features & Use Cases

  • Full Configuration System Reference: Covers global and project-level config locations, dot-notation access patterns, the /config modal, pricing registry setup, and plugin config widgets.
  • Event Bus & Hook Guidance: Documents both Python async hooks and no-code JSON hooks for intercepting lifecycle events, tool calls, permissions, and user input, with working examples for logging, blocking, modifying payloads, and context injection.
  • Extension Workflow Support: Provides step-by-step guidance for building Python plugins, registering slash commands, extending the unified tool pipeline, adhering to stdout and render state rules, creating dynamic system prompts, and writing passing integration tests.
  • Use Case: A developer can use this Skill to add a custom JSON hook that blocks shell commands containing rm -rf during work hours, or build a Python plugin that injects the current git branch and status into every LLM request.

Quick Start

Use the kollabor-harness skill to add a custom hook that logs all user prompts submitted in Kollab to a local audit file for compliance tracking.

Frequently Asked Questions about kollabor-harness

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

FAQPage Schema
How do I add Python plugins to my terminal AI workspace?

To add Python plugins to your terminal AI workspace, use this Skill's extension workflow guidance for building plugins, registering slash commands, and extending the unified tool pipeline. It provides step-by-step instructions for adhering to render state rules and passing integration tests.

Can I intercept tool calls and user input with JSON lifecycle hooks?

Yes, you can intercept tool calls and user input using no-code JSON lifecycle hooks. This Skill documents how to register hooks on the event bus for logging, blocking, modifying payloads, and injecting context without writing Python code.

What is the best way to modify global and project-level configuration in Kollab?

The best way to modify global and project-level configuration is through the config system using dot-notation access patterns and the config modal. This Skill provides reference locations for setting up pricing registries and plugin config widgets.

Does the Kollab harness support blocking specific shell commands during automated workflows?

Yes, the Kollab harness supports blocking specific shell commands during automated workflows. You can build a custom JSON hook to intercept and block commands, such as preventing unsafe operations during work hours.

How do I create dynamic system prompts for an event-driven AI plugin?

To create dynamic system prompts for an event-driven AI plugin, follow this Skill's guidance on extending the architecture. It covers injecting context like git branch status into LLM requests and adhering to render state best practices.

Why write integration tests for Kollab features and slash commands?

Writing integration tests for Kollab features and slash commands ensures your custom plugins function correctly within the event bus. This Skill provides testing spec creation guidance to validate your tool pipeline extensions.