gini

Official

Tune your agent via /api/* recipes

AuthorLilac-Labs
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill helps users configure, extend, and operate Gini’s runtime correctly by grounding self-directed changes in the exact /api/* endpoints and registered tool behaviors instead of misleading CLI habits.

Core Features & Use Cases

  • API-first self-configuration: Explains that Gini operates through /api/* and the registered tool catalog, with the CLI treated as a human-operator mirror rather than an agent mechanism.
  • Practical “recipe book” workflows: Provides common operational playbooks for browser sign-in/connect cycles, scheduled jobs, messaging bridge setup (Telegram), MCP server registration, connectors, and delegated subagents.
  • Safety and approvals guidance: Clarifies approval seams, risk classifications, and how approval mode (strict, auto, yolo) affects high-risk tool execution so users can prompt accurately without false denials.
  • Operational troubleshooting: Includes targeted recovery steps for common runtime failures such as Telegram health errors, missing Chromium, managed-mode browser launch issues, and stuck high-risk approvals.

Quick Start

Ask the agent to explain the correct /api/* call and tool path for a specific capability you want to configure or operate (for example, how to set up a Telegram bridge or connect the managed browser profile).

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: gini
Download link: https://github.com/Lilac-Labs/gini-agent/archive/main.zip#gini

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.