using-kli-cli

Access KLI task graph and MCP tools via Bash commands.

26|3|Updated Jul 3, 2026
One-click install
npx skills add https://github.com/kleisli-io/kli --skill using-kli-cli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: using-kli-cli
Source: https://github.com/kleisli-io/kli/tree/main/plugin/skills/using-kli-cli
Command: npx skills add https://github.com/kleisli-io/kli --skill using-kli-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables interaction with the KLI task graph using the command line, providing a crucial interface when graphical tools or direct MCP access are unavailable, ensuring task management continuity.

Core Features & Use Cases

  • Task Graph Interaction: Manage and query tasks, observations, and patterns directly from a Bash environment.
  • Teammate Collaboration: Facilitates task graph access for AI agents acting as teammates.
  • Use Case: When an AI teammate needs to record a finding or check the status of a task without direct access to the main KLI interface, it can use these Bash commands.

Quick Start

Use the kli observe command to record a new finding for a specific task.

Frequently Asked Questions about using-kli-cli

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

FAQPage Schema
How do I interact with a task graph from the command line when direct MCP access is unavailable?

To manage KLI tasks via Bash, use the kli binary to expose MCP tools as subcommands, enabling task querying and management for AI agents and users without direct MCP access.

How do I record an observation for a specific task using Bash?

To record an observation for a specific task using Bash, execute the `kli observe` command, which allows you to log new findings directly into the task graph from your terminal environment.

Can an AI agent teammate query task statuses without accessing the main KLI interface?

Yes, an AI agent teammate can query task statuses and record findings by executing kli Bash commands, accessing the task graph and MCP tools independently of the main graphical KLI interface.

Does this command line approach support task context management for AI agents?

Yes, this command line approach supports task context management by leveraging the kli binary to expose MCP tools as subcommands, ensuring continuous task observation and management for AI agents in Bash.

What are the limitations of managing task graphs through Bash commands?

Managing task graphs through Bash commands is limited to environments where the kli binary is installed, serving as a fallback interface when graphical tools or direct MCP access are unavailable.