ericlitman/open-pstack

pstack for Claude Code and Codex, tracking Cursor pstack upstream

What it solves

Open Pstack provides a structured engineering framework for coding agents, moving them beyond simple code generation to a rigorous software development lifecycle. It solves the problem of agents producing unreliable code by enforcing engineering rules, step-by-step workflows, and a requirement for real-world verification rather than just passing tests.

How it works

It functions as a plugin for coding agents (specifically for Claude Code and Codex). It does not introduce a new model but instead provides the agent with a set of focused skills and playbooks. The primary entry point is poteto-mode, which routes tasks through a specific process: understanding the existing system, comparing designs, implementing small changes, and using multiple models to challenge decisions.

Who it’s for

Developers using Claude Code or Codex who want to implement high-velocity, high-reliability agentic coding workflows based on the methodologies used at Cursor.

Highlights

  • Multi-Model Review: Can orchestrate several models (e.g., Opus, Sol, Grok) in parallel to interrogate designs and break diffs.
  • Verification-First Approach: Prioritizes checking real behavior in the application over relying solely on test suites.
  • Specialized Engineering Skills: Includes dedicated tools for architectural design (architect), system explanation (how/why), and handling CI failures (babysit).
  • Evidence-Based Trust: Encourages a supervised start where agents must leave inspectable evidence of their work before being granted more autonomy.

Related

  • Project
    feiskyer/claude-code-settings

    A set of Claude Code plugins (skills & sub‑agents) that add brainstorming, deep research, GitHub issue/PR automation, image generation, translation, and more, with ready‑made config files for many LLM providers.

  • Project
    ooiyeefei/ccc

    A collection of custom plugins and skills for Claude Code that adds specialized capabilities for product management, agentic system design, and automated technical workflows.

  • Project
    levnikolaevich/claude-code-skills

    A collection of standalone agent skills and plugins for Claude Code and OpenAI Codex that provide structured, professional engineering workflows across the entire software development lifecycle.

  • Project
    michaelshimeles/skills

    A curated set of Claude Code agent skills that automate common development tasks—visual diff generation, service‑layer structuring, evidence‑driven UI testing, iterative Greptile review loops, work‑tree isolation, and prose clean‑up—plus a ready‑to‑drop workflow tying them together.

  • Project
    DannyMac180/skills

    A collection of reusable AI agent skills and plugins for Claude Code and Codex, enabling advanced supervised workflows, mod management, and personalized learning.