# Skynet Brain Lab

`src/skynet` is the experimental brain lab for `OpenSkyNet`, not the platform itself.

The separation should stay explicit:

- `OpenSkyNet` = runtime, channels, tools, memory, validation, autonomy substrate
- `Omega` = internal control/runtime line inside the platform
- `Skynet Brain Lab` = search for a new cognitive substrate beyond a plain LLM-centric agent

This repo should be operated under a two-line directive:

1. `OpenSkyNet`
   Keep the platform solid, measurable, and operational.
2. `Skynet Brain Lab`
   Search for a new brain, new substrate, and more general cognition than the current architecture provides.

The lab is allowed to be more radical than the platform.
The platform is not required to mirror the lab.

Current working posture:

- `OpenSkyNet` is in relative stabilization mode
- only continuity or operational bug fixes should touch the platform for now
- new architecture work should happen here first

## Why This Exists

`OpenSkyNet` is already useful and relatively solid as an operational agent.
That makes it a bad place to invent an entirely new brain in-line.

`src/skynet` exists to do the opposite:

- test non-traditional substrates
- study limits of the current LLM-centric architecture
- run ugly or high-risk experiments without destabilizing the platform
- distill mechanisms small enough to survive falsification and later promotion

## What Belongs Here

### 1. Brain Search

Architectures that try to go beyond:

- next-token dependency
- flat textual memory
- purely heuristic control
- traditional monolithic nets

Examples:

- bifasic / cyborg architectures
- adaptive local plasticity
- surprise-gated compute
- spectral / oscillatory memory
- topology-changing or field-based substrates

### 2. Hypothesis Extraction

Not "which old version wins".
Instead:

- which mechanism is valuable
- in what task should it win
- what cost it adds
- whether it transfers to runtime reality

Examples:

- `adaptive_decay`
- `surprise_gated_compute`
- `runtime observers`
- `causal valence`

### 3. Promotion Gate

Nothing should move into `Omega` or the platform just because it sounds deep.

A lab result should only be promoted when:

1. the mechanism is isolated
2. the benchmark is falsable
3. the gain is real against a relevant baseline
4. the runtime integration is smaller than the added complexity
5. the platform still works without `src/skynet`

## Current Layout

- `doc/`
  Theory, papers, and conceptual roadmaps. Use as hypothesis fuel, not as proof.
- `analysis/`
  Brain Lab analysis, architecture audits, benchmark readings, and next-cycle decisions.
- `experiments/`
  One-off runnable probes, historical lines, and benchmark scripts.
- `runtime-observer/`
  Learning from real runtime traces.
- `causal-valence/`
  Structured causal labeling and world-transition inference.
- `cognitive-kernel/`
  Lightweight online learners trained from trajectory data.
- `artifacts/`
  Reports, replay utilities, and derived evidence.
- `lab/`
  Reserved area for future cleaner lab surfaces; currently underused.

## Working Rule

If the goal is:

- make `OpenSkyNet` more reliable or cheaper -> work in platform / `Omega`
- discover a new mind topology -> work here first

If a result is promising but still fragile:

- keep it in the lab
- design a benchmark where it should win on its own terms
- only then ask whether it transfers into the platform

The lab should be free to fail.
The platform should not pay for those failures prematurely.

## Priority Research Lines

### Line A: Cheap Fast Path + Rare Deep Path

Best current signal from old `EX` work.
This is where `V67`-style surprise gating has already produced transferable wins.

### Line B: Local Plasticity / Adaptive Decay

Still alive, but not yet strong enough for kernel promotion.
Needs tasks closer to real continuity and interruption handling.

### Line C: Spectral / Oscillatory Memory

Interesting, but currently under-validated.
Needs tasks where phase, periodicity, or latent continuity can actually win.

### Line D: Cyborg / Biphasic Substrates

Best treated as a research family, not as one giant file to import.
The useful unit is likely a mechanism, not a whole architecture.

## Anti-Pattern

Do not use `src/skynet` as:

- a dumping ground for mythology
- a hidden dependency of the kernel
- a second copy of `Omega`
- a narrative justification for unvalidated complexity

The lab is valuable only if it finds mechanisms the platform would not have discovered on its own.