skills.rest
llmh333
llmh333COMMUNITY
Explore all skills by this author →
GitHub

autonomous-agent-harness

Automate scheduled Gemini CLI agent workflows with persistent memory and tool execution.

Updated Sep 13, 2025
Search Tags:#dispatch#autonomous agent#persistent memory#computer use#MCP tools#scheduled tasks#task queue
One-click install
npx skills add https://github.com/llmh333/employee_management_spring --skill autonomous-agent-harness-llmh333
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autonomous-agent-harness
Source: https://github.com/llmh333/employee_management_spring/tree/main/.gemini/skills/autonomous-agent-harness
Command: npx skills add https://github.com/llmh333/employee_management_spring --skill autonomous-agent-harness-llmh333

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of building and operating continuous or scheduled AI agents that can remember across sessions and reliably execute multi-step workflows with tool support.

Core Features & Use Cases

  • Persistent Memory: Uses Gemini CLI memory plus an MCP memory server for cross-session structured knowledge and retrieval.
  • Scheduled Operations (Crons): Creates recurring autonomous tasks using native Gemini CLI crons to run prompts on a schedule in isolated sessions.
  • Tool-Driven Autonomy: Supports dispatch-driven remote agent runs, MCP tool integration, and optional computer use for browser/desktop automation.
  • Task Queuing: Maintains a persistent task queue via memory files so work survives session boundaries.

Use Case: Set up a “daily PR reviewer” that checks repositories on a cron, reviews changes, posts results, and writes outcomes back to memory for future context.

Quick Start

Ask for an always-on assistant setup by saying: “Create a scheduled daily agent at 9am that reviews my open pull requests, summarizes findings to memory, and flags issues for follow-up.”

Frequently Asked Questions about autonomous-agent-harness

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

FAQPage Schema
How do I set up scheduled autonomous agents that maintain persistent memory across sessions?▼

You can set up scheduled autonomous agents with persistent memory by using Gemini CLI crons to run prompts on a schedule and an MCP memory server to store cross-session structured knowledge. This enables recurring tasks to retain context over time.

Can I trigger remote agent workflows on demand instead of using a fixed cron schedule?▼

Yes, you can trigger remote agent workflows on demand using dispatch-based triggering. This allows operational tasks to execute as needed outside of scheduled crons, running multi-step workflows with MCP tools within a self-directing loop.

What is the best way to automate a daily PR review agent that writes results back to memory?▼

The best way to automate a daily PR review agent is using scheduled crons combined with persistent memory files. The agent runs at specified times, reviews repository changes, and writes outcomes back to memory for future context and follow-up.

Does Gemini CLI support computer use and browser automation for autonomous agents?▼

Yes, Gemini CLI supports optional computer use for browser and desktop automation within autonomous agents. This integrates with MCP tools and dispatch triggering to allow self-directing loops to interact with graphical interfaces.

How do autonomous agents handle task queuing when work survives session boundaries?▼

Autonomous agents handle task queuing by maintaining a persistent task queue via memory files. This ensures that scheduled and dispatch-triggered operational tasks survive session boundaries and resume processing in subsequent isolated sessions.

Related Skills

Explore More Software Engineering Skills→
hlibkovalhlibkoval
1

agent-sdk-doc

Build autonomous agents with Python, TypeScript, MCP, hooks, and observability.

Community
Advanced
auldsyababuaauldsyababua
6

personal-assistant

Manage schedules, tasks, and reminders with persistent memory.

Community
Advanced
zero3041zero3041
1

autonomous-agent-harness

Configure Claude Code with persistent memory, scheduled tasks, and computer use.

Community
Advanced
himanshu231204himanshu231204
2

autonomous-agent-harness

Configure persistent self-directing agent workflows with Claude Code and MCP servers.

Community
Advanced