Productivity

Ralph loop implementation with agent mode for sustained autonomous task completion

The Ralph Wiggum loop technique: simplified autonomous agent loop with minimal setup

Create JSON-format task lists for ralph-tui

Create Beads tasks (git-backed, with dependencies) for ralph-tui

Generate a structured prd.json task list for use with ralph-tui's autonomous loop

Branch close checklist: tests, commit message, pull request, and review request

Use git worktrees to run parallel agent sessions on separate branches

Split work across parallel subagents and coordinate their outputs

Orchestrate specialized subagents for different parts of a task

Prepare code for review: self-review, test coverage, and pull request description

Execute a plan step-by-step with checkpoints and verification at each stage

Write structured implementation plans before starting complex tasks

Hypothesis-driven debugging loop: observe, hypothesize, test, verify

Structured ideation and problem decomposition frameworks

Create, test, and publish new skills from within your agent

Full browser automation: navigate, click, fill forms, extract data, and screenshot

Discover and install skills from skills.sh directly inside an agent session

Master Vibe Coder Workflows 2026: Build, Author & Automate with Claude AI

Codex skill for WeChat Mini Program engineering workflows

Zero-setup Codex Skill: one skill for Claude CLI multi-agent collaboration

Shared Codex skills for SLS and other workflows

Codex skill for maintaining repo-level project memory files.

A Codex skill for checkpointing and resuming long-running tasks across conversations.

Agentic Lib is a language-agnostic starting point for agentic software workflows. It provides reusable Codex skill instructions for chaining discovery, planning, implementation, and review while keeping context small through progressive disclosure.