kinic-cli

Operate kinic-cli memory canister workflows with correct auth and network routing.

17|2|Updated Nov 3, 2025
One-click install
npx skills add https://github.com/ICME-Lab/kinic-cli --skill kinic-cli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kinic-cli
Source: https://github.com/ICME-Lab/kinic-cli/tree/main/skills
Command: npx skills add https://github.com/ICME-Lab/kinic-cli --skill kinic-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Kinic CLI turns day-to-day operations of verifiable, owned memory canisters into clear, repeatable commands, reducing confusion about which flags, auth mode, and network choice to use.

Core Features & Use Cases

  • Auth mode selection for agents and humans: Choose identity vs Internet Identity, and understand which commands require which kind of auth.
  • Network-aware command routing: Decide when to use mainnet flags vs local replica defaults, including when to ask the user for network assumptions before mutating state.
  • Discovery and retrieval-focused workflows: Use machine-readable capabilities to discover supported operations, then use list/show/search to inspect and retrieve memories, optionally via JSON for agent-friendly consumption.
  • Local preference management: Treat prefs as a local JSON API for shared CLI and TUI behavior, including default memory and retrieval tuning.
  • TUI entrypoint for interactive ops: Launch a terminal UI when a human wants a guided flow rather than automation.

Quick Start

Ask the user for which identity and network (local vs mainnet), then run the smallest read command to inspect their memories (for example: cargo run -- --ic --identity alice list --json).

Frequently Asked Questions about kinic-cli

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

FAQPage Schema
How do I manage trustless memory canisters from the command line?

You manage trustless memory canisters by selecting correct auth modes and network routing flags to run discovery, listing, showing, and searching operations. You choose between identity and Internet Identity authentication before executing commands.

How do I get JSON output for agent workflows on the Internet Computer?

JSON output for agent workflows is supported specifically for list, show, search, capabilities, and preferences commands. You apply the JSON flag to these supported operations to retrieve machine-readable memory canister data for automated consumption.

Do I need Internet Identity to search memory canisters?

Internet Identity is not required for searching memory canisters. You can execute discovery, listing, showing, and searching as read-only operations using local replica defaults or mainnet flags without needing identity authentication.

What's the best way to configure local preferences for terminal UI memory workflows?

Local preferences for terminal UI memory workflows are configured as a local JSON API. You manage default memory and retrieval tuning settings that are shared across both CLI commands and TUI interactive entrypoints.

When should I use mainnet flags versus local replica defaults for canister operations?

Mainnet flags versus local replica defaults are chosen based on your target network for canister operations. You should ask the user for network assumptions before mutating state, using the mainnet flag for production and local defaults for development.

Can I discover supported operations for memory canisters without reading documentation?

You can discover supported operations for memory canisters without reading documentation by using machine-readable capabilities. This capability discovery command returns available operations programmatically before you attempt listing, showing, or searching memories.