Railspike — AI skill for Claude Code
Keep your AI coding agents on track.
How to install Railspike
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open gulkily/railspike and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Railspike does
Keep your AI coding agents on track. Railspike leads to clear requirements, plans you approve, and atomic implementation stages with verification and a commit trail. Plain markdown, one git command to install, tested extensively with Codex, Claude Code, and others. Your agents write the code, you call the shots.
Alternatives in AI
- Flow Define — Multi-AI requirements scoping using Codex and Gemini CLIs (Double Diamond Define phase) 2.8k ★
- Maestro Companion — Quick execution for small tasks — minimal run lifecycle (start + done) with evidence recording 530 ★
- MioIsland — macOS Dynamic Island for AI coding agents 388 ★
README
Railspike
**Make your coding agent write the plan down first.**
Railspike is a set of Markdown instructions your agent follows. The agent writes requirements and a staged plan into your repo, then stops until you approve each one. After that it works on a branch, one stage at a time, verifying and committing each stage with a record of what changed.
Use it with Claude Code, Codex, GitHub Copilot, or another agent that can read your project files. Railspike has no CLI or runtime dependencies; installation and updates use Git.
[Website](https://gulkily.github.io/railspike/) · [RAILSPIKE.md](RAILSPIKE.md) · [Step files](docs/workflow/)
Why I built this
I wanted plans that live in the repo, that I can edit in my own editor and come back to the next day, with explicit review points when the direction changes.
So Railspike puts every plan in a Markdown file. You can read it in VS Code, Obsidian, vim, or a GitHub diff. You can edit it directly or ask the agent to change it. The instructions require the agent to stop until you type `Approved Step N`. These are instructions the agent follows, not a technical enforcement mechanism.
Use Railspike alongside your agent’s plan mode when you want to keep decisions in editable files across sessions. Compare approaches, combine the parts that fit, and revise the requirements and stages before approving the work.
You shape the plan
The agent's first draft is a starting point. Before you approve anything, you can change it two ways:
- Edit the file yourself. Delete a requirement, rewrite a stage, add a note for the agent. Then tell it to reread the file.
- Ask the agent to revise it. "Drop the caching stage." "Make the error message less technical." "What happens if the file is empty?"
You can mix both in the same step.
Step 1 is where this matters most. The agent lays out a few approaches, and you don't have to pi
Related Skills
Memory Trail
Decision memory and session logging for AI-assisted development. Track WHY decisions are made, not just WHAT c
PM Leads
A lightweight, streamlined approach for product managers and sales teams to track customer engagement directly
Finpilot
AI agent that turns bank statements into clear insights: categories, subscriptions, upcoming bills, unusual sp
ASD SDE 80 Karpathy
Agent skill that makes AI answers clear: short sentences, plain words, one idea per sentence (80% ASD-STE100).
Loomwatch
Put your AI apps to work as a team. Connect Claude Code, Codex and OpenCode on one canvas, ask in plain words,
PayPilot
Chat-first AI billing agent. Describe an invoice in plain language, approve it, and PayPilot creates and sends
Related Agents
Lead Judge
Use to work the Lead Radar queue: read each lead, agree or disagree with its score, approve the real ones with
Standup Poster
Writes clear team updates and posts them to Slack — always after you approve the wording.
Crew Engineer
cgp-crew engineer (Opus, high effort). Carries out delicate work whose approach is already clear: refactors ac