triz-analysis

TRIZ Analysis — Rust Edition

Systematic innovation audit for Rust codebases, workspace architectures, and CI/CD pipelines.

Input Requirements

• Target Scope: The specific directory, pipeline, or framework to analyze (e.g., crates/, scripts/, .github/workflows/, agent-coordination).
• Language Focus: Rust-specific patterns (ownership, async, traits, lifetimes).

Output Convention

All agent-generated analysis must be written to analysis/ using:

analysis/triz-<scope>-YYYY-MM-DD.md

Example: analysis/triz-scripts-2026-06-18.md

Audit Process

1. SCOPE SCAN: Examine the target scope for existing pain points or trade-offs.
2. CONTRADICTION DISCOVERY: Identify technical contradictions: "Improving [X] in this scope causes [Y] to worsen."
3. PRINCIPLE MAPPING: Map discovered contradictions to TRIZ inventive principles.
4. INNOVATION ROADMAP: Propose specific Rust-idiomatic changes based on TRIZ resolutions.
5. REPORT: Document findings in analysis/.

Core Protocol

Contradiction-First Approach

1. SCAN the scope
2. IDENTIFY contradiction: "Improving [X] causes [Y] to worsen"
3. CHECK if contradiction is real or apparent
4. RESOLVE using separation principles
5. DOCUMENT findings in analysis/

Ideal Final Result (IFR) for Rust Audits

"The ideal Rust system performs its function with minimal unsafe code, unnecessary cloning, and contention"

1. What is the ideal outcome for this scope?
2. What prevents reaching it? (current type/ownership/CI constraints)
3. Which constraints are real vs assumed?
4. Could the borrow checker enforce correctness instead of runtime checks?

Contradiction Matrix (Rust-Specific)

Improving	vs Worsening	→ Principle	Rust Pattern
Throughput	vs Latency	#7, #13, #17	Batching, bounded queues, workload-aware scheduling
API ergonomics	vs Binary size	#1, #15	Feature flags, #[cfg], trait-based DI
Compile time	vs Runtime perf	#13, #35	LTO off, cargo-bloat profile tuning
Safety	vs Performance	#12, #4	unsafe minimization, scoped SIMD
Genericity	vs Compile time	#2, #6	Carefully scoped generics, sealed traits
Async surface	vs Blocking cost	#1, #15	Workload-aware direct execution, spawn_blocking, bounded workers
Throughput	vs Fairness	#13, #15	Batching, bounded poll work, deliberate yield_now
Concurrency	vs Resource usage	#1, #35	Semaphore, bounded mpsc, worker pools
Shared state	vs Contention	#1, #2	Ownership/message passing, short lock scope
Build matrix	vs CI minutes	#1, #9	cargo-hack, workspace inheritance
Test coverage	vs Test runtime	#13, #35	cargo-nextest sharding, proptest shrinking