Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/XxYouDeaDPunKxX/Signal-Rail/llms.txt

Use this file to discover all available pages before exploring further.

Signal Rail is a plain-text document governance system for personal and AI-assisted projects. It gives every piece of project material a destination — a specific file with a specific role — so nothing gets buried, promoted too early, or confused with something it is not. Think of it as a set of rails that keep your project readable while it grows, branches, changes shape, and comes back after interruption.

Introduction

Learn what Signal Rail is, why it exists, and the problem it solves for project material.

Quickstart

Bootstrap a Signal Rail instance in minutes using the Windows installer or manual copy.

The Rails

Explore every canonical rail — from runtime entry to archive — and what belongs in each.

AI Agent Operation

Understand how AI agents activate the lateral kernel and operate safely inside an instance.

How Signal Rail works

Signal Rail prevents the most common failure in project notes: everything living in the same undifferentiated space. A note becomes a decision. A possibility becomes a constraint. A current blocker gets buried under old context. Signal Rail fixes this by asking one practical question before any material is written:
What kind of material is this, and where should it live?
1

Bootstrap an instance

Run init_signal_rail.bat (Windows) or copy the canonical files manually into the folder you want to govern. The instance is a copy of Signal Rail living beside or inside your real project.
2

Open with an AI agent

Ask your AI to read 00_runtime_entry.txt, then 06_ai_to_ai.txt. This activates the lateral kernel and establishes the operating frame before any action is taken.
3

Close the working frame

The agent closes the minimum frame: host project, working object, active mode, source scope, and authority. No writing happens until the frame is closed.
4

Route material to the right rail

Current live work goes to 03_master_working.txt. Decisions already won go to 04_decision_log.txt. Strong-but-unresolved ideas go to 05_latent_ideas.txt. Everything has a home.

The thirteen rails

Signal Rail separates material by role across thirteen canonical and lateral files.

Runtime Entry (00)

The entry gate. Read before anything else.

Orientation (01)

Project identity, perimeter, and reading frame.

Protocol Freeze (02)

Hard-to-reopen identity constants.

Master Working (03)

Current live state, blockers, and next move.

Decision Log (04)

Decisions already taken and already in effect.

Latent Ideas (05)

Live unresolved material awaiting a home.

Lateral Kernel (06)

Governs AI agent behaviour inside an instance.

Guided Prompts (07)

Safer guided paths for starts, rebuilds, reviews, and routing passes.

Surface Map (08)

Technical topology, entrypoints, and minimal runbook.

Handoff (09)

Continuity and re-entry between sessions.

Field Findings (97)

Temporary lateral capture during an active pass before routing or discard.

Parking (98)

Useful but not active material held safely outside canonical rails.

Archive (99)

Closed, historical, or no-longer-live material.

Key concepts

Core Concepts

Rails, levels, authority, and promotion — the foundational ideas behind Signal Rail.

Guided Prompts

Structured question flows for safe session starts, source rebuilds, and idea intake.

Routing & Promotion

How material moves between rails without blurring level or authority.

File Map

Quick-reference table of every Signal Rail file and its canonical role.

Build docs developers (and LLMs) love