OthmanAdi / planning-with-files

TL;DR

AI coding agents often struggle with context loss during long-running tasks, leading to inefficiencies. The tool 'planning-with-files' was developed to create persistent, markdown-based plans that remain intact even after interruptions.

✦ Why It Matters

Engineers can leverage persistent planning to improve the reliability and collaboration of AI coding agents in their projects.

Key Takeaways

Implement the planning-with-files skill for better task continuity.
Utilize markdown files to track phases, findings, and progress.
Reduce session recovery time significantly with this skill.
Ensure your agent retains context across resets and crashes.

How It Works

The planning-with-files skill operates by creating three dedicated markdown files that store the agent's task plan, findings, and progress logs. This information is re-injected into the agent's context at the start of each turn, allowing it to maintain continuity even after context resets.

Lifecycle hooks automate the process of updating these files, ensuring that important information is consistently captured and utilized.

OthmanAdi / planning-with-files

tailwindlabs / tailwindcss

denoland / deno

angular / angular

Third-party cyber evaluations involving OpenAI models

tailwindlabs / tailwindcss

denoland / deno

angular / angular

Third-party cyber evaluations involving OpenAI models