Rules

cursorkleosr

This project provides a streamlined way to work with AI assistants (like Claude or GPT-4) inside the Cursor IDE, making development more autonomous and consistent. It helps the AI remember project context and follow a structured process, even across different sessions. Think of it as giving your AI assistant a reliable memory and a clear playbook.

Upstream ↗Seed list ↗Role: workflow-framework
Why it matters

Profile

This project provides a streamlined way to work with AI assistants (like Claude or GPT-4) inside the Cursor IDE, making development more autonomous and consistent. It helps the AI remember project context and follow a structured process, even across different sessions. Think of it as giving your AI assistant a reliable memory and a clear playbook.

setup mediumintegration mediuminterface editor
Provenance

Signals

Listed in the awesome-cursor-ecosystem seed README

Sources: 2 / Surfaces: 1

Fast skim

What the upstream surface says

Short excerpt only, so you can decide whether to click out.

Two markdown files at the repo root carry project memory and live task state for Cursor. The idea is to give the assistant something stable to read on startup instead of re-deriving goals and stack from chat each time.

project_config.md holds slow-changing facts: goals, stack, patterns, constraints, token notes.

workflow_state.md holds the moving parts: current phase, plan, task table, metrics, checkpoints, log. The model (or you) is expected to update this as work moves.

cursorkleosrFilesSetupPhasesDiagramBlueprints and gitLicenseContributing
  • Edit project_config.md for your project.
  • Set phase and status under the dynamic block in workflow_state.md if you are starting from scratch.
  • In Cursor, ask the assistant to read both files before acting and to write results back to workflow_state.md.