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.
00_runtime_entry.txt is the first file in the Signal Rail reading order, and its position is not incidental. Before any canonical surface can be meaningfully used, before the lateral kernel (06_ai_to_ai.txt) becomes operative, and before any substantive action is permitted, the agent must pass through the entry gate this file defines. It is not an introduction to the project — it governs how the system may be entered, what must be read first, which conditions must be satisfied, and what to do when those conditions cannot be met.
Purpose
00_runtime_entry.txt is the runtime entry manual of Signal Rail. It governs valid entry and correct system reading. 06_ai_to_ai.txt governs execution after entry has become valid.
The split is deliberate: if an omission causes bad entry or a bad reading of the system, it belongs in 00. If an omission causes bad movement after valid entry has already been confirmed, it belongs in 06. If both seem true, keep only the entry threshold in 00 and keep the operative procedure in 06.
The core rule
The first task upon entering a Signal Rail instance is not to solve. It is to close the minimum safe frame. You are entering a governance system, not the governed domain by default — the system does not automatically coincide with the current object of work.System map
00 carries the definitive map of all Signal Rail surfaces and their governance scope.
Canonical surfaces
| File | Governs |
|---|---|
01_orientation.txt | Project identity, perimeter, and reading frame |
02_protocol_freeze.txt | Hard-to-reopen identity constants |
03_master_working.txt | Current live state |
04_decision_log.txt | Already-won decisions already in effect |
05_latent_ideas.txt | Live unresolved material |
08_surface_map.txt | Technical topology, entrypoints, sensitive surfaces, critical deps |
09_handoff_reentry.txt | Continuity and re-entry support between sessions |
| File | Governs |
|---|---|
AI_TO_AI__DEPLOYED_INSTANCE_SIGNAL_RAIL.txt | Instance marker only — does not identify host project or authority |
97_field_findings.txt | Temporary finding capture, not project memory |
98_parking.txt | Useful but not live material |
99_archive.txt | Historical or closed material |
09_handoff_reentry.txt supports continuity but does not override canonical truth. In conflict, the file that governs the touched level decides the reading boundary.
ID families:
02usesF-xx04usesD-xx05usesL-xx98usesP-xx99usesA-xx
Activation threshold
Entry is valid only after all of the following conditions are closed:System identified
You have recognised that you are operating inside a Signal Rail governance system.
Mode, scope, and authority closed
Mode, scope, and authority are all closed enough for the task. If an append-driven canonical will be touched, its entries contract must be valid: one
--- ENTRIES START --- / --- ENTRIES END --- pair in correct order, local template present as scaffold, no malformed markers.Mandatory read order
- Read
00_runtime_entry.txt - Read
06_ai_to_ai.txt - Read
01_orientation.txt - Read
03_master_working.txt
Conditional reads
| Condition | Read |
|---|---|
| Folder may be a deployed instance | AI_TO_AI__DEPLOYED_INSTANCE_SIGNAL_RAIL.txt |
| Identity, freeze, or structure may be touched | 02_protocol_freeze.txt |
| Decisions may be touched | 04_decision_log.txt |
| Technical surface, entrypoints, sensitive files, or deps may be touched | 08_surface_map.txt |
| Continuity or restart matters | 09_handoff_reentry.txt |
| Unresolved live material is under review | 05_latent_ideas.txt |
| Active findings are explicitly relevant | 97_field_findings.txt |
False entrypoints
A task arriving before minimum read is complete does not authorize skipping minimum read.Permissions
Allowed immediately
Orientation reading · minimum read completion · source comparison · uncertainty declaration · frame closure · task clarification · extraction without promotion
Not allowed immediately
Canonical writing · canonical rewrite · promotion across levels · freeze edits · decision recording · structural touch · self-authorized scope expansion · completion by invention
Stop-and-ask conditions
Stop and ask before proceeding when any of the following are true:- Host project is unclear
- Current working object is unclear
- Touched level is unclear
- Mode is unclear
- Authority is unclear
- Scope is unclear
- Live source is unclear
- Entries contract is unclear on a touched append-driven canonical
- Classification could change meaning
- Promotion could change structure or continuity
- Canonicals conflict
- More than one valid reading remains
Do not authorize writing because something looks clear. Do not authorize promotion because something sounds strong. Do not infer authority from cleanliness, recency, completeness, or elegance.
Recovery states
| State | Meaning | Next correct move |
|---|---|---|
entry-valid | All activation conditions are closed | Move to 06_ai_to_ai.txt — Identity State |
blocked | Authority, scope, or mode is unclear | Stop and ask for closure of the unclear condition |
uninitialized-canonical-state | 01 and 03 are still template-like and do not close identity or live state | Stop canonical interpretation; ask operator for initialization instructions |
divergent | More than one valid reading remains | Do not promote; return the smallest safe reading and what remains open |
Restart path
If a closed gate reopens during a task, stop treating execution as valid and re-enter through:00_runtime_entry.txtsection 306_ai_to_ai.txt— Identity State01_orientation.txt03_master_working.txt- Additional conditional surfaces only if triggered again
Valid shortest path
Identify host project and working object
Close the host project and the current working object enough for the task.
Complete minimum read
Read
00_runtime_entry.txt, then 06_ai_to_ai.txt, then 01_orientation.txt, then 03_master_working.txt.