Efficiency Layer Roadmap

Reference documentation for Efficiency Layer Roadmap in Adaptive Skills.

Goal

Turn the Efficiency Layer into a clear, bounded roadmap for improving how work is carried out with Adaptive Skills.

This roadmap is not about changing the current Evolution Layer baseline. It is about reducing waste in:

• task framing
• context growth
• oversized work slices
• weak checkpoints
• poor handoffs
• overpowered default model use

Current position

The Efficiency Layer currently exists as a documented thesis, not as a published skill set.

That is intentional. The first step should remain docs-first so the repository can separate:

• a library problem
• a workflow problem
• an orchestration problem
• an AletheIA concern

v0 — framing and operating patterns

The v0 phase should document the operating patterns that make work lighter without over-process.

It should establish:

• what efficiency problems are in scope
• what signals show a task is too large or too vague
• what a healthy checkpoint looks like
• what should remain in AletheIA or project-local overlays instead of entering this library

Expected output of v0:

• a bounded problem map
• first candidate skill briefs
• explicit boundaries versus Evolution Layer and AletheIA

v0.1 — first candidate skills

The first candidate skills are:

• task-chunking
• handoff-summary
• checkpoint-review

They are first because they:

• are easy to validate through real usage
• address repeated operational waste
• have clearer boundaries than more agent-like concerns
• can work with or without AletheIA

Deferred for later

These candidates should remain deferred until the first three are validated:

• context-compression
• prompt-framing
• model-routing

They have higher overlap risk with:

• model behavior
• orchestration policy
• AletheIA-style macro guidance

Boundary rules

Efficiency Layer should not:

• reopen the Evolution Layer v1.1 baseline
• become a hidden AletheIA dependency
• optimize only for token count
• replace normal feature, design, or quality skills

Efficiency Layer should:

• reduce avoidable rework
• make checkpoints cheaper and clearer
• improve handoff discipline
• help teams keep work slices bounded

Recommended next implementation step

After this roadmap pass, the next sensible move is a separate documentation pass that defines:

• one brief per candidate skill
• expected outputs
• where each skill stops and AletheIA begins
• one lightweight example of combined usage

First pilot artifacts

The pilot artifacts for this first use pass are:

• docs/efficiency-layer-first-pilot.md
• docs/efficiency-layer-pilot-checklist.md
• templates/pilot/efficiency-pilot-report.md
• docs/efficiency-layer-crisis-monitor-reference.md
• examples/efficiency/crisis-monitor-efficiency-reference.md
• docs/efficiency-layer-trio-patterns.md
• docs/efficiency-layer-next-signals.md