Pilot Shell is a tool for running Claude Code and Codex with a spec-driven workflow. It is described as helping move from requirement to production-grade code, with planned, tested, verified output.
Its core workflow is /spec, which is used to plan, approve, implement, and verify work. The page also describes TDD enforcement, where tests come before implementation, plus quality hooks that auto-format, lint, and type-check on every edit. Other listed features include persistent memory that carries context across sessions, semantic code search that finds code by meaning rather than keywords, pre-wired MCP and LSP servers for both agents, and worktree isolation for safe parallel work via git worktrees.
Pilot Shell is presented for engineers working with Claude Code and Codex. The installation instructions show a terminal-based installer run with curl and bash, after which claude or codex can be run in any project. The page also mentions generating rules with /setup-rules and shipping features with /spec.
Pilot Shell is a CLI tools & terminal project. It focuses on automating and enforcing best practices for agent-driven code generation and testing using Claude Code and Codex. It is built as an open-source project for software engineers and developers using AI coding agents. The project is open source (Open Source). Pilot Shell is available on the command line, and it can be self-hosted.
Behind Pilot Shell is Max Ritter, and it first shipped in 2025. The project is developed in the open on GitHub with 1.8k stars and 228 commits in the last 90 days. Key capabilities include spec-driven workflow, TDD enforcement, and quality hooks. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do